Publications

@inproceedings{conf-ipdps-jiajun24,
  title="{An Optimized Error-controlled MPI Collective Framework Integrated with Lossy Compression}",
  author={Huang, Jiajun and Di, Sheng and Yu, Xiaodong and Zhai, Yujia and Zhang, Zhaorui and Liu, Jinyang and Lu, Xiaoyi and Raffenetti, Ken and Zhou, Hui and Zhao, Kai and Chen, Zizhong and Cappello, Franck and Guo, Yanfei and Thakur, Rajeev},
  booktitle={Proceedings of the 38th IEEE International Parallel and Distributed Processing Symposium (IPDPS)},
  publisher = {IEEE Computer Society},
  series = {IPDPS '24},
  abstract = {With the ever-increasing computing power of supercomputers and the growing scale of scientific applications, the efficiency of MPI collective communications turns out to be a critical bottleneck in large-scale distributed and parallel processing. The large message size in MPI collectives is particularly concerning because it can significantly degrade the overall parallel performance. To address this issue, prior research simply applies the off-the-shelf fix-rate lossy compressors in the MPI collectives, leading to suboptimal performance, limited generalizability, and unbounded errors. In this paper, we propose a novel solution, called C-Coll, which leverages error-bounded lossy compression to significantly reduce the message size, resulting in a substantial reduction in communication cost. The key contributions are three-fold. (1) We develop two general, optimized lossy-compression-based frameworks for both types of MPI collectives (collective data movement as well as collective computation), based on their particular characteristics. Our framework not only reduces communication cost but also preserves data accuracy. (2) We customize SZx, an ultra-fast error-bounded lossy compressor, to meet the specific needs of collective communication. (3) We integrate C-Coll into multiple collectives, such as MPI_Allreduce, MPI_Scatter, and MPI_Bcast, and perform a comprehensive evaluation based on real-world scientific datasets. Experiments show that our solution outperforms the original MPI collectives as well as multiple baselines and related efforts by 1.8-2.7X.},
  year={2024},
  address={San Francisco, California, USA},
  month={May 27-31}
}

@article{journals-micro24-liuyao,
author = {Dai, Liuyao and Qi, Hao and Chen, Weicong and Lu, Xiaoyi},
title = "{High-Speed Data Communication with Advanced Networks in Large Language Model Training}",
journal = {IEEE Micro},
year = {2024},
volume = {44},
number = {02},
issn = {1937-4143},
pages = {31-40},
abstract = {Large language models (LLMs) like Generative Pre-trained Transformer, Bidirectional Encoder Representations from Transformers, and T5 are pivotal in natural language processing. Their distributed training is influenced by high-speed interconnects. This article characterizes their training performance across various interconnects and communication protocols: TCP/IP, Internet Protocol over InfiniBand, (IPoIB), and Remote Direct Memory Access (RDMA), using data and model parallelism. RDMA-100 Gbps outperforms IPoIB-100 Gbps and TCP/IP-10 Gbps, with average gains of 2.5x and 4.8x in data parallelism, while in model parallelism, the gains were 1.1x and 1.2x. RDMA achieves the highest interconnect utilization (up to 60 Gbps), compared to IPoIB with up to 20 Gbps and TCP/IP with up to 9 Gbps. Larger models demand increased communication bandwidth, with AllReduce in data parallelism consuming up to 91% of training time, and forward receive and back-embedding AllReduce in model parallelism taking up to 90%. The larger-scale experiment confirms that communication predominates iterations. Our findings underscore the significance of communication in distributed LLM training and present opportunities for optimization.},
keywords = {training;parallel processing;data models;computational modeling;decoding;tcpip;synchronization;high-speed networks;large language models},
doi = {10.1109/MM.2024.3360081},
publisher = {IEEE Computer Society},
address = {Los Alamitos, CA, USA},
series = {IEEE Micro'24},
month = {mar}
}

@article{journals-micro24-yuke,
author = {Li, Yuke and Kashyap, Arjun and Guo, Yanfei and Lu, Xiaoyi},
title = "{Compression Analysis for BlueField-2/3 Data Processing Units: Lossy and Lossless Perspectives}",
journal = {IEEE Micro},
year = {2024},
volume = {44},
number = {02},
issn = {1937-4143},
pages = {8-19},
abstract = {A data processing unit (DPU) with programmable smart network interface card containing system-on-chip (SoC) cores is now a valuable addition to the host CPU, finding use in high-performance computing (HPC) and data center clusters for its advanced features, notably, a hardware-based data compression engine (C-engine). With the convergence of big data, HPC, and machine learning, data volumes burden communication and storage, making efficient compression vital. This positions DPUs as tools to accelerate compression workloads and enhance data-intensive applications. This article characterizes lossy (e.g., SZ3) and lossless (e.g., DEFLATE, lz4, and zlib) compression algorithms using seven real-world datasets on Nvidia BlueField-2/-3 DPUs. We explore the potential opportunities for offloading these compression workloads from the host. Our findings demonstrate that the C-engine within the DPU can achieve up to 26.8x speedup compared to its SoC core. We also provide insights on harnessing BlueField for compression, presenting seven crucial takeaways to steer future compression research with DPUs.},
keywords = {engines;task analysis;data compression;throughput;memory management;hardware acceleration;distributed databases;data processing;smart devices;system-on-chip},
doi = {10.1109/MM.2023.3343636},
publisher = {IEEE Computer Society},
address = {Los Alamitos, CA, USA},
series = {IEEE Micro'24},
month = {mar}
}

@inproceedings{conf-ipdps24-yuke,
  title="{Accelerating Lossy and Lossless Compression on Emerging BlueField DPU Architectures}",
  author={Li, Yuke and Kashyap, Arjun and Chen, Weicong and Guo, Yanfei and Lu, Xiaoyi},
  booktitle={Proceedings of the 38th IEEE International Parallel and Distributed Processing Symposium (IPDPS)},
  publisher = {IEEE Computer Society},
  series = {IPDPS '24},
  year={2024},
  address={San Francisco, California, USA},
  note={Best Paper Award Nomination},
  month={May 27-31}
}

@inproceedings{conf-ics24-jiajun-gzccl,
  author    = {Huang, Jiajun and Di, Sheng and Yu, Xiaodong and Zhai, Yujia and Liu, Jinyang and Huang, Yafan and Raffenetti, Ken and Zhou, Hui and Zhao, Kai and Lu, Xiaoyi and Chen, Zizhong and Cappello, Franck and Guo, Yanfei and Thakur, Rajeev},
  title     = "{gZCCL: Compression-Accelerated Collective Communication Framework for GPU Clusters}",
  booktitle = {Proceedings of the 38th International Conference on Supercomputing (ICS)},
  year      = {2024},
  address  = {Kyoto University, Kyoto, Japan},
  month     = {June},
  series = {ICS '24},
  abstract = {GPU-aware collective communication has become a major bottleneck for modern computing platforms as GPU computing power rapidly rises. A traditional approach is to directly integrate lossy compression into GPU-aware collectives, which can lead to serious performance issues such as underutilized GPU devices and uncontrolled data distortion. In order to address these issues, in this paper, we propose gZCCL, a first-ever general framework that designs and optimizes GPU-aware, compression-enabled collectives with an accuracy-aware design to control error propagation. To validate our framework, we evaluate the performance on up to 512 NVIDIA A100 GPUs with real-world applications and datasets. Experimental results demonstrate that our gZCCL-accelerated collectives, including both collective computation (Allreduce) and collective data movement (Scatter), can outperform NCCL as well as Cray MPI by up to 4.5X and 28.7X, respectively. Furthermore, our accuracy evaluation with an image-stacking application confirms the high reconstructed data quality of our accuracy-aware framework.}
}

@inproceedings{conf-ipdps-jasanur24,
  title="{DRUTO: Upper-Bounding Silent Data Corruption Vulnerability in GPU Applications}",
  author={Rahman, Md Hasanur and Di, Sheng and Guo, Shengjian and Lu, Xiaoyi and Li, Guanpeng and Cappello, Franck},
  booktitle={Proceedings of the 38th IEEE International Parallel and Distributed Processing Symposium (IPDPS)},
  publisher = {IEEE Computer Society},
  series = {IPDPS '24},
  year={2024},
  address={San Francisco, California, USA},
  month={May 27-31}
}

@inproceedings{conf-ipdps24-darren,
  title="{NVMe-oPF: Designing Efficient Priority Schemes for NVMe-over-Fabrics with Multi-Tenancy Support}",
  author={Ng, Darren and Lin, Andrew and Kashyap, Arjun and Li, Guanpeng and Lu, Xiaoyi},
  booktitle={Proceedings of the 38th IEEE International Parallel and Distributed Processing Symposium (IPDPS)},
  publisher = {IEEE Computer Society},
  series = {IPDPS '24},
  year={2024},
  address={San Francisco, California, USA},
  month={May 27-31}
}

@inproceedings{conf-hoti-llm-hao23,
  author={Qi, Hao and Dai, Liuyao and Chen, Weicong and Jia, Zhen and Lu, Xiaoyi },
  title="{Performance Characterization of Large Language Models on High-Speed Interconnects}",
  location={ ONLINE },
  booktitle = {Proceedings of the 30th IEEE Hot Interconnects Symposium},
  publisher = {IEEE Computer Society},
  series = {HotI '23},
  month={ August },
  year = 2023
}

@inproceedings{conf-modsim-liuyao23,
    author={Dai, Liuyao and Weingram, Adam and Qi, Hao, and Chen, Weicong and Lu, Xiaoyi},
    title={{LogGOPSGauger: A Work-In-Progress Tool for Gauging LogGOPS Model with GPU-Aware Communication}},
    booktitle={2023 Workshop on Modeling \& Simulation of Systems and Applications (ModSim)},
    location={Seattle, WA, USA},
    month={ August },
    series = {ModSim '23},
    note = {Poster Paper},
    year={2023}
}

@inproceedings{conf-mlsysw-yujie23,
title="{Learning Distributed Protocols with Zero Knowledge}",
author={Hui, Yujie and Ripberger, Drew and Lu, Xiaoyi and Wang, Yang},
booktitle={Machine Learning for Systems 2023 (MlSys Workshop at NeurIPS 2023)},
year={2023},
series = {MlSys Workshop'23},
url={https://openreview.net/forum?id=u0Ncut8ru5}
}

@inproceedings{conf-ipdps-chen23,
  author = {Chen, Weicong and Qi, Hao and Lu, Xiaoyi and Tatsuoka, Curtis},
  booktitle = {Proceedings of IEEE International Parallel and Distributed Processing Symposium},
  keywords = {Group testing, Bayesian, Lattices, Spark, COVID-19},
  publisher = {IEEE Computer Society},
  title = {SBGT: Scaling Bayesian-based Group Testing for Disease Surveillance},
  abstract = {The COVID-19 pandemic underscored the necessity for disease surveillance using group testing. Novel Bayesian methods using lattice models were proposed, which offer substantial improvements in group testing efficiency by precisely quantifying uncertainty in diagnoses, acknowledging varying individual risk
and dilution effects, and guiding optimally convergent sequential pooled test selections using a Bayesian Halving Algorithm. Computationally, however, Bayesian group testing poses considerable challenges as computational complexity grows exponentially with sample size. This can lead to shortcomings in reaching a desirable scale without practical limitations. We propose a new framework for scaling Bayesian group testing based on Spark: SBGT. We show that SBGT is lightning fast and highly scalable. In particular, SBGT is up to 376x, 1733x, and 1523x faster than the state-of-the-art framework in manipulating lattice models, performing test selections, and conducting statistical analyses, respectively, while achieving up to 97.9\% scaling efficiency up to 4096 CPU cores. More importantly, SBGT fulfills our mission towards reaching applicable scale for guiding pooling decisions in wide-scale disease surveillance, and other large scale group testing applications.},
  series = {IPDPS '23},
  month = {May},
  year = {2023}
}

@article{journals-jcst-Weingram23,
author = {Weingram, Adam and Li, Yuke and Qi, Hao and Ng, Darren and Dai, Liuyao and Lu, Xiaoyi},
title = {xCCL: A Survey of Industry-Led Collective Communication Libraries for Deep Learning},
publisher = {Journal of Computer Science and Technology},
year = {2023},
journal = {Journal of Computer Science and Technology},
volume = {38},
number = {1},
eid = {166},
numpages = {29},
pages = {166},
keywords = {collective; deep learning; distributed training; GPUDirect; RDMA (remote direct memory access)},
url = {https://jcst.ict.ac.cn/EN/abstract/article_2965.shtml},
doi = {10.1007/s11390-023-2894-6},
abstract = {Machine learning techniques have become ubiquitous both in industry and academic applications. Increasing model sizes and training data volumes necessitate fast and efficient distributed training approaches. Collective communications greatly simplify inter- and intra-node data transfer and are an essential part of the distributed training process as information such as gradients must be shared between processing nodes. In this paper, we survey the current state-of-the-art collective communication libraries (namely xCCL, including NCCL, oneCCL, RCCL, MSCCL, ACCL, and Gloo), with a focus on the industry-led ones for deep learning workloads. We investigate the design features of these xCCLs, discuss their use cases in the industry deep learning workloads, compare their performance with industry-made benchmarks (i.e., NCCL Tests and PARAM), and discuss key take-aways and interesting observations. We believe our survey sheds light on potential research directions of future designs for xCCLs.},
note = {Invited Paper for the Special Issue in Honor of Professor Kai Hwang’s 80th Birthday},
series = {JCST '23}
}    

@inproceedings{conf-sc-hao23-poster,
author = {Qi, Hao and Dai, Liuyao and Chen, Weicong and Lu, Xiaoyi},
title = "{Early Experience in Characterizing Training Large Language Models on Modern HPC Clusters}",
month = {Nov.},
date = {12-17},
year = {2023},
publisher = {Association for Computing Machinery},
address = {New York, NY, USA},
booktitle = {Proceedings of the International Conference for High Performance Computing, Networking, Storage and Analysis},
keywords = {LLM, RDMA, Performance},
location = {Denver, Colorado},
note = {Research Poster Paper},
series = {SC '23}
}

@inproceedings{conf-sc-yuke23-poster,
author = {Li, Yuke and Guo, Yanfei and Lu, Xiaoyi},
title = "{Characterizing One-/Two-sided Designs in OpenSHMEM Collectives}",
month = {Nov.},
date = {12-17},
year = {2023},
publisher = {Association for Computing Machinery},
address = {New York, NY, USA},
booktitle = {Proceedings of the International Conference for High Performance Computing, Networking, Storage and Analysis},
keywords = {OpenSHMEM, PGAS, Omni-Path, Performance},
location = {Denver, Colorado},
note = {Research Poster Paper},
series = {SC '23}
}

@inproceedings{conf-modsim-ben23,
    title={{Early Experiences in Modeling Performance Implications of DPU-Offloaded Computation}},
    author={Chen, Weicong and Li, Yuke and Kashyap, Arjun and Lu, Xiaoyi},
    booktitle={2023 Workshop on Modeling \& Simulation of Systems and Applications (ModSim)},
    location={Seattle, WA, USA},
    month={ August },
    series = {ModSim '23},
    year={2023}
}

@inproceedings{conf-sc-darren23-poster,
author = {Ng, Darren and Parkinson, Charles and Lin, Andrew and Kashyap, Arjun and  Lu, Xiaoyi},
title = "{An Early Case Study with Multi-Tenancy Support in SPDK’s NVMe-over-Fabric Designs}",
month = {Nov.},
date = {12-17},
year = {2023},
publisher = {Association for Computing Machinery},
address = {New York, NY, USA},
booktitle = {Proceedings of the International Conference for High Performance Computing, Networking, Storage and Analysis},
keywords = {Storage Disaggregation, NVMe-over-Fabric, SPDK},
location = {Denver, Colorado},
note = {Research Poster Paper},
series = {SC '23}
}

@inproceedings{conf-hoti-dpu-yuke23,
  author={Li, Yuke and Kashyap, Arjun and Guo, Yanfei and Lu, Xiaoyi},
  title="{Characterizing Lossy and Lossless Compression on Emerging BlueField DPU Architectures}",
  location={ ONLINE },
  booktitle = {Proceedings of the 30th IEEE Hot Interconnects Symposium},
  publisher = {IEEE Computer Society},
  series = {HotI '23},
  month={ August },
  year = 2023
}

@inproceedings{conf-hotinfra-tianxi23,
    title={{On the Discontinuation of Persistent Memory: Looking Back to Look Forward}},
    author={Li, Tianxi and Wang, Yang and Lu, Xiaoyi},
    booktitle={The 1st International Workshop on Hot Topics in System Infrastructure (HotInfra), in conjunction with International Symposium on Computer Architecture (ISCA)},
    location={Orlando, FL, USA},
    month={ June },
    date={18},
    url={https://hotinfra23.github.io/papers/hotinfra23-paper9.pdf},
    series = {HotInfra '23},
    year={2023}
}

@book{book-hpbdc-mit22,
place = {Cambridge, MA}, 
title = "{High-Performance Big Data Computing}", 
publisher = {The MIT Press}, 
author = {Panda, Dhabaleswar K. and Lu, Xiaoyi and Shankar, Dipti}, 
year = {2022},
abstract = {Over the last decade, the exponential explosion of data known as big data has changed the way we understand and harness the power of data. The emerging field of high-performance big data computing, which brings together high-performance computing (HPC), big data processing, and deep learning, aims to meet the challenges posed by large-scale data processing. This book offers an in-depth overview of high-performance big data computing and the associated technical issues, approaches, and solutions.
The book covers basic concepts and necessary background knowledge, including data processing frameworks, storage systems, and hardware capabilities; offers a detailed discussion of technical issues in accelerating big data computing in terms of computation, communication, memory and storage, codesign, workload characterization and benchmarking, and system deployment and management; and surveys benchmarks and workloads for evaluating big data middleware systems. It presents a detailed discussion of big data computing systems and applications with high-performance networking, computing, and storage technologies, including state-of-the-art designs for data processing and storage systems. Finally, the book considers some advanced research topics in high-performance big data computing, including designing high-performance deep learning over big data (DLoBD) stacks and HPC cloud technologies.},
isbn = {9780262046855},
url="https://mitpress.mit.edu/9780262046855/high-performance-big-data-computing/",
series="MIT Press '22"
}

@article{conf-vldb-yang22,
author = {Wang, Yang and Yu, Miao and Hui, Yujie and Zhou, Fang and Huang, Yuyang and Zhu, Rui and Ren, Xueyuan and Li, Tianxi and Lu, Xiaoyi},
title = {A Study of Database Performance Sensitivity to Experiment Settings},
year = {2022},
issue_date = {March 2022},
publisher = {VLDB Endowment},
volume = {15},
number = {7},
issn = {2150-8097},
url = {https://doi.org/10.14778/3523210.3523221},
doi = {10.14778/3523210.3523221},
abstract = {To allow performance comparison across different systems, our community has developed multiple benchmarks, such as TPC-C and YCSB, which are widely used. However, despite such effort, interpreting and comparing performance numbers is still a challenging task, because one can tune benchmark parameters, system features, and hardware settings, which can lead to very different system behaviors. Such tuning creates a long-standing question of whether the conclusion of a work can hold under different settings.This work tries to shed light on this question by reproducing 11 works evaluated under TPC-C and YCSB, measuring their performance under a wider range of settings, and investigating the reasons for the change of performance numbers. By doing so, this paper tries to motivate the discussion about whether and how we should address this problem. While this paper does not give a complete solution---this is beyond the scope of a single paper, it proposes concrete suggestions we can take to improve the state of the art.},
journal = {Proc. VLDB Endow.},
month = {mar},
pages = {1439–1452},
numpages = {14},
series = {VLDB '22},
pdf = {https://www.vldb.org/pvldb/vol15/p1439-wang.pdf}
}

@article{journals-CoRR-shashank22,
  author    = {Shashank Gugnani and
               Scott Guthridge and
               Frank Schmuck and
               Owen Anderson and
               Deepavali Bhagwat and
               Xiaoyi Lu},
  title     = {Arcadia: {A} Fast and Reliable Persistent Memory Replicated Log},
  journal   = {CoRR},
  volume    = {abs/2206.12495},
  year      = {2022},
  url       = {https://doi.org/10.48550/arXiv.2206.12495},
  doi       = {10.48550/arXiv.2206.12495},
  eprinttype = {arXiv},
  series = {arXiv '22},
  eprint    = {2206.12495},
  timestamp = {Mon, 04 Jul 2022 16:22:19 +0200},
  biburl    = {https://dblp.org/rec/journals/corr/abs-2206-12495.bib},
  bibsource = {dblp computer science bibliography, https://dblp.org}
}

@inproceedings{conf-hpdc-arjun22,
author = {Kashyap, Arjun and Lu, Xiaoyi},
title = {NVMe-oAF: Towards Adaptive NVMe-oF for IO-Intensive Workloads on HPC Cloud},
year = {2022},
isbn = {9781450391993},
publisher = {Association for Computing Machinery},
address = {New York, NY, USA},
url = {https://doi.org/10.1145/3502181.3531476},
doi = {10.1145/3502181.3531476},
abstract = {Applications running inside containers or virtual machines, traditionally use TCP/IP for communication in HPC clouds and data centers. The TCP/IP path usually becomes a major performance bottleneck for applications performing NVMe-over-Fabrics (NVMe-oF) based I/O operations in disaggregated storage settings. We propose an adaptive communication channel, called NVMe-over-Adaptive-Fabric (NVMe-oAF), that applications could leverage to eliminate the high-latency and low-bandwidth incurred by remote I/O requests over TCP/IP. NVMe-oAF accelerates I/O intensive applications using locality awareness along with optimized shared memory and TCP/IP paths. The adaptiveness of the fabric stems from the ability to adaptively select shared memory or TCP channel and further applying optimizations for the chosen channel. To evaluate NVMe-oAF, we co-design Intel's SPDK library with our designs and show up to 7.1x bandwidth improvement and up to 4.2x latency reduction for various workloads over commodity TCP/IP-based Ethernet networks (e.g., 10Gbps, 25Gbps, and 100Gbps). We achieve similar (or sometimes better) performance when compared to NVMe-over-RDMA by avoiding the cumbersome management of RDMA in HPC cloud environments. Finally, we also co-design NVMe-oAF with H5bench to showcase the benefit it brings to HDF5 applications. Our evaluation indicates up to a 7x bandwidth improvement when compared with the network file system (NFS).},
booktitle = {Proceedings of the 31st International Symposium on High-Performance Parallel and Distributed Computing},
pages = {56–70},
numpages = {15},
keywords = {nvme-over-fabrics, shared memory, hpc cloud, spdk},
location = {Minneapolis, MN, USA},
pdf = {https://dl.acm.org/doi/pdf/10.1145/3502181.3531476},
series = {HPDC '22}
}

@inproceedings{conf-hipc-Chen22,
  author = {Chen, Weicong and Lu, Xiaoyi and Tatsuoka, Curtis},
  booktitle = {Proceedings of the 29th IEEE International Conference on High Performance Computing},
  publisher = {IEEE},
  title = {HiBGT: High-Performance Bayesian Group Testing for COVID-19},
  series = {HiPC '22},
  note = {Acceptance Rate: 26\%, 35/131},
  pdf = {https://sites.ucmerced.edu/files/luxi/files/hipc22-cr.pdf},
  year = 2022
}

@article{journals-biostatistics-ben22,
    author = {Tatsuoka, Curtis and Chen, Weicong and Lu, Xiaoyi},
    title = "{Bayesian Group Testing with Dilution Effects}",
    journal = {Biostatistics},
    year = {2022},
    month = {04},
    abstract = "{A Bayesian framework for group testing under dilution effects has been developed, using lattice-based models. This work has particular relevance given the pressing public health need to enhance testing capacity for coronavirus disease 2019 and future pandemics, and the need for wide-scale and repeated testing for surveillance under constantly varying conditions. The proposed Bayesian approach allows for dilution effects in group testing and for general test response distributions beyond just binary outcomes. It is shown that even under strong dilution effects, an intuitive group testing selection rule that relies on the model order structure, referred to as the Bayesian halving algorithm, has attractive optimal convergence properties. Analogous look-ahead rules that can reduce the number of stages in classification by selecting several pooled tests at a time are proposed and evaluated as well. Group testing is demonstrated to provide great savings over individual testing in the number of tests needed, even for moderately high prevalence levels. However, there is a trade-off with higher number of testing stages, and increased variability. A web-based calculator is introduced to assist in weighing these factors and to guide decisions on when and how to pool under various conditions. High-performance distributed computing methods have also been implemented for considering larger pool sizes, when savings from group testing can be even more dramatic.}",
    issn = {1465-4644},
    doi = {10.1093/biostatistics/kxac004},
    url = {https://doi.org/10.1093/biostatistics/kxac004},
    note = {kxac004},
    eprint = {https://academic.oup.com/biostatistics/advance-article-pdf/doi/10.1093/biostatistics/kxac004/43355708/kxac004.pdf},
    series = {Biostatistics '22}    
}

@article{journals-tbench-yukeli22,
title = {Understanding hot interconnects with an extensive benchmark survey},
journal = {BenchCouncil Transactions on Benchmarks, Standards and Evaluations},
volume = {2},
number = {3},
pages = {100074},
year = {2022},
issn = {2772-4859},
doi = {https://doi.org/10.1016/j.tbench.2022.100074},
url = {https://www.sciencedirect.com/science/article/pii/S2772485922000618},
author = {Yuke Li and Hao Qi and Gang Lu and Feng Jin and Yanfei Guo and Xiaoyi Lu},
series = {TBench '22},
keywords = {Benchmarks, Interconnects, RDMA},
abstract = {Understanding the designs and performance characterizations of hot interconnects on modern data center and high-performance computing (HPC) clusters is a fruitful research topic in recent years. The rapid and continuous growth of high-bandwidth and low-latency communication requirements for various types of data center and HPC applications (such as big data, deep learning, and microservices) has been pushing the envelope of advanced interconnect designs. We believe this is high time to investigate the performance characterizations of representative hot interconnects with different benchmarks. Hence, this paper presents an extensive survey of state-of-the-art hot interconnects on data center and HPC clusters and the associated representative benchmarks to help the community to better understand modern interconnects. In addition, we characterize these interconnects by the related benchmarks under different application scenarios. We provide our perspectives on benchmarking data center interconnects based on our survey, experiments, and results.}
}

@inproceedings{conf-bench-darren22,
	Address = {Online},
	Author = {Ng, Darren and Schmierer, Colin and Lin, Andrew and Liu, Zeyu and Yu, Falin and Newsam, Shawn and Ehsani, Reza and Lu, Xiaoyi},
	Booktitle = {Proceedings of International Symposium on Benchmarking, Measuring, and Optimizing},
	Da = {2022//},
	Publisher = {Springer International Publishing},
	Title = {Benchmarking Object Detection Models with Mummy Nuts Dataset},
	Ty = {CONF},
	series = {Bench '22},
	video = {https://www.youtube.com/watch?v=tjaZa7lZq3Q&ab_channel=BenchCouncil},
	Year = {2022}
}

@inproceedings{conf-ipdps-shashank21,
  author = {Gugnani, Shashank and Li, Tianxi and Lu, Xiaoyi},
  Date-Added = {2021-02-22 19:42:43 +0000},
  Date-Modified = {2021-02-03 04:02:36 +0000},
  booktitle = {Proceedings of IEEE International Parallel and Distributed Processing Symposium},
  crossref = {conf/ipdps/2021},
  keywords = {Checkpoint/Restart, NVMe, NVMf, Exascale},
  publisher = {IEEE Computer Society},
  title = {NVMe-CR: A Scalable Ephemeral Storage Runtime for Checkpoint/Restart with NVMe-over-Fabrics},
  abstract = {Emerging SSDs with NVMe-over-Fabrics (NVMf) support provide new opportunities to significantly improve the performance of IO-intensive HPC applications. However, state-of-the-art parallel filesystems can not extract the best possible performance from fast NVMe SSDs and are not designed for latency-critical ephemeral IO tasks, such as checkpoint/restart. In this paper, we propose a powerful abstraction called microfs to peel away unnecessary software layers and eliminate namespace coordination. Building upon this abstraction, we present the design of NVMe-CR, a scalable ephemeral storage runtime for clusters with disaggregated compute and storage. NVMe-CR proposes techniques like metadata provenance, log record coalescing, and logically isolated shared device access, built around the microfs abstraction, to reduce the overhead of writing millions of concurrent checkpoint files. NVMe-CR utilizes high-density all-flash arrays accessible via NVMf to absorb bursty checkpoint IO and increase the progress rates of HPC applications obliviously. Using the ECP CoMD application as a use case, results show that on a local cluster our runtime can achieve near perfect (> 0.96) efficiency at 448 processes. Moreover, our designs can reduce checkpoint overhead by as much as 2x compared to state-of-the- art storage systems.},
  series = {IPDPS '21},
  year = {2021}
}

@inproceedings{conf-words-ms,
  added-at = {2021-04-16T00:00:00.000+0200},
  author = {Lu, Xiaoyi and Kashyap, Arjun},
  booktitle = {The Second Workshop On Resource Disaggregation and Serverless (WORDS’21), co-located with ASPLOS 2021},
  crossref = {conf/words/2021},
  keywords = {Offloadable, Migratable, Microservice, Disaggregated Architectures},
  publisher = {ACM},
  timestamp = {2021-06-25T11:40:11.000+0200},
  title = {Towards Offloadable and Migratable Microservices on Disaggregated Architectures: Vision, Challenges, and Research Roadmap},
  url = {https://wuklab.github.io/words/words21-lu.pdf},
  note = {Vision Paper},
  series = {WORDS '21},
  year = 2021
}

@inproceedings{conf-hpdc-dstore,
  added-at = {2021-06-25T00:00:00.000+0200},
  author = {Gugnani, Shashank and Lu, Xiaoyi},
  booktitle = {Proceedings of International ACM Symposium on High Performance and Distributed Computing},
  crossref = {conf/hpdc/2021},
  keywords = {Tailless, Quiescent-Free, Object Store, PMEM},
  publisher = {ACM},
  timestamp = {2021-06-25T11:40:11.000+0200},
  title = {DStore: A Fast, Tailless, and Quiescent-Free Object Store for PMEM},
  series = {HPDC '21},
  year = 2021
}

@article{conf-vldb-shashank21,
	Author = {Gugnani, Shashank and Kashyap, Arjun and Lu, Xiaoyi},
	Date-Added = {2021-01-26 19:42:43 +0000},
	Date-Modified = {2021-02-03 04:02:36 +0000},
	Journal = {Proceedings of the VLDB Endowment},
	Number = {4},
	Pages = {626--639},
	Title = {Understanding the Idiosyncrasies of Real Persistent Memory},
	Abstract = {High capacity persistent memory (PMEM) is finally commercially available in the form of Intel’s Optane DC Persistent Memory Module (DCPMM). Researchers have raced to evaluate and understand the performance of DCPMM itself as well as systems and applications designed to leverage PMEM resulting from over a decade of research. Early evaluations of DCPMM show that its behavior is more nuanced and idiosyncratic than previously thought. Several assumptions made about its performance that guided the design of PMEM-enabled systems have been shown to be incorrect. Unfortunately, several peculiar performance characteristics of DCPMM are related to the memory technology (3D-XPoint) used and its internal architecture. It is expected that other technologies (such as STT-RAM, memristor, ReRAM, NVDIMM), with highly variable characteristics, will be commercially shipped as PMEM in the near future. Current evaluation studies fail to understand and categorize the idiosyncratic behavior of PMEM; i.e., how do the peculiarities of DCPMM related to other classes of PMEM. Clearly, there is a need for a study which can guide the design of systems and is agnostic to PMEM technology and internal architecture.
In this paper, we first list and categorize the idiosyncratic behavior of PMEM by performing targeted experiments with our proposed PMIdioBench benchmark suite on a real DCPMM platform. Next, we conduct detailed studies to guide the design of storage systems, considering generic PMEM characteristics. The first study guides data placement on NUMA systems with PMEM while the second study guides the design of lock-free data structures, for both eADR- and ADR-enabled PMEM systems. Our results are often counter-intuitive and highlight the challenges of system design with PMEM.},
	Volume = {14},
	series = {VLDB '21},
	pdf = {http://vldb.org/pvldb/vol14/p626-gugnani.pdf},
        code = {https://github.com/padsys/PMIdioBench},
	Year = {2021}}

@inproceedings{conf-sc-tianxi21,
	Author = {Li, Tianxi and Shi, Haiyang and Lu, Xiaoyi},
	Booktitle = {Proceedings of the International Conference for High Performance Computing, Networking, Storage and Analysis},
	Location = {St. Louis, Missouri},
	Publisher = {Association for Computing Machinery},
	Address = {New York, NY, USA},
	Series = {SC '21},
	Title = {HatRPC: Hint-Accelerated Thrift RPC over RDMA},
	Year = {2021}
}

@INPROCEEDINGS{conf-sec-yujie21,
  author={Hui, Yujie and Lien, Jeffrey and Lu, Xiaovi},
  booktitle={2021 IEEE/ACM Symposium on Edge Computing (SEC)}, 
  title={Characterizing and Accelerating End-to-End EdgeAI Inference Systems for Object Detection Applications}, 
  year={2021},
  volume={},
  number={},
  pages={01-12},
  doi={10.1145/3453142.3491294},
  series = {SEC '21}  
}

@inproceedings{conf-bench-arjun20,
	Address = {Cham},
	Author = {Kashyap, Arjun and Gugnani, Shashank and Lu, Xiaoyi},
	Booktitle = {Proceedings of International Symposium on Benchmarking, Measuring, and Optimizing},
	Da = {2020//},
	Date-Added = {2021-01-26 19:45:38 +0000},
	Date-Modified = {2021-02-03 04:03:02 +0000},
	Editor = {Gao, Wanling and Zhan, Jianfeng and Wolf, Felix},
	Publisher = {Springer International Publishing},
	Title = {Impact of Commodity Networks on Storage Disaggregation with NVMe-oF},
	Ty = {CONF},
	series = {Bench '20},
	Year = {2020}}

@inproceedings{conf-sc-tianxi20,
	Abstract = {Byte-addressable persistent memory (PMEM) can be directly manipulated by Remote Direct Memory Access (RDMA) capable networks. However, existing studies to combine RDMA and PMEM can not deliver the desired performance due to their PMEM-oblivious communication protocols. In this paper, we propose novel PMEM-aware RDMA-based communication protocols for persistent key-value stores, referred to as <u>R</u>emote <u>D</u>irect <u>M</u>emory <u>P</u>ersistence based <u>K</u>ey-<u>V</u>alue stores (RDMP-KV). RDMP-KV employs a hybrid 'server-reply/server-bypass' approach to 'durably' store individual key-value objects on PMEM-equipped servers. RDMP-KV's runtime can easily adapt to existing (server-assisted durability) and emerging (appliance durability) RDMA-capable interconnects, while ensuring server scalability through a lightweight consistency scheme. Performance evaluations show that RDMP-KV can improve the server-side performance with different persistent key-value storage architectures by up to 22x, as compared with PMEM-oblivious RDMA-'Server-Reply' protocols. Our evaluations also show that RDMP-KV outperforms a distributed PMEM-based filesystem by up to 65% and a recent RDMA-to-PMEM framework by up to 71%.},
	Articleno = {52},
	Author = {Li, Tianxi and Shankar, Dipti and Gugnani, Shashank and Lu, Xiaoyi},
	Booktitle = {Proceedings of the International Conference for High Performance Computing, Networking, Storage and Analysis},
	Date-Added = {2021-01-26 19:35:59 +0000},
	Date-Modified = {2021-02-03 04:03:50 +0000},
	Isbn = {9781728199986},
	Keywords = {next generation networking, data storage systems, persistent memory, key-value stores, RDMA},
	Location = {Atlanta, Georgia},
	Publisher = {IEEE Press},
	Series = {SC '20},
	Title = {RDMP-KV: Designing Remote Direct Memory Persistence Based Key-Value Stores with PMEM},
	Year = {2020}}

@article{journals-jcst-JinSHZL20,
author = {Jin, Zheng-Hao and Shi, Haiyang and Hu, Ying-Xin and Zha, Li and Lu, Xiaoyi},
title = {CirroData: Yet Another SQL-on-Hadoop Data Analytics Engine with High Performance},
publisher = {Journal of Computer Science and Technology},
year = {2020},
journal = {Journal of Computer Science and Technology},
volume = {35},
number = {1},
eid = {194},
numpages = {14},
pages = {194},
keywords = {CirroData;high performance;SQL-on-Hadoop;online analytical processing (OLAP);Big Data},
url = {http://jcst.ict.ac.cn/EN/abstract/article_2598.shtml},
series = {JCST '20},
doi = {10.1007/s11390-020-9536-z}
}    

@inproceedings{conf-sc-haiyang20,
author = {Shi, Haiyang and Lu, Xiaoyi},
title = {INEC: Fast and Coherent in-Network Erasure Coding},
year = {2020},
isbn = {9781728199986},
publisher = {IEEE Press},
abstract = {Erasure coding (EC) is a promising fault tolerance scheme that has been applied to many well-known distributed storage systems. The capability of Coherent EC Calculation and Networking on modern SmartNICs has demonstrated that EC will be an essential feature of in-network computing. In this paper, we propose a set of coherent in-network EC primitives, named INEC. Our analyses based on the proposed α-β performance model demonstrate that INEC primitives can enable different kinds of EC schemes to fully leverage the EC offload capability on modern SmartNICs. We implement INEC on commodity RDMA NICs and integrate it into five state-of-the-art EC schemes. Our experiments show that INEC primitives significantly reduce 50th, 95th, and 99th percentile latencies, and accelerate the end-to-end throughput, write, and degraded read performance of the key-value store co-designed with INEC by up to 99.57%, 47.30%, and 49.55%, respectively.},
booktitle = {Proceedings of the International Conference for High Performance Computing, Networking, Storage and Analysis},
articleno = {66},
numpages = {17},
keywords = {erasure coding, fault tolerance, in-network computing, next generation networking},
location = {Atlanta, Georgia},
series = {SC '20}
}

@inproceedings{conf-bench-HuiLL19,
  added-at = {2020-06-10T00:00:00.000+0200},
  author = {Hui, Yujie and Lien, Jeffrey and Lu, Xiaoyi},
  biburl = {https://www.bibsonomy.org/bibtex/27a7c44a3dfc8cd83e0e596e900dd609d/dblp},
  booktitle = {Proceedings of International Symposium on Benchmarking, Measuring, and Optimizing},
  crossref = {conf/bench/2019},
  editor = {Gao, Wanling and Zhan, Jianfeng and Fox, Geoffrey C. and Lu, Xiaoyi and Stanzione, Dan},
  ee = {https://doi.org/10.1007/978-3-030-49556-5_3},
  interhash = {30850ff0979cee822cc832913ddbd41d},
  intrahash = {7a7c44a3dfc8cd83e0e596e900dd609d},
  isbn = {978-3-030-49556-5},
  keywords = {dblp},
  pages = {32-48},
  publisher = {Springer},
  timestamp = {2020-06-11T11:40:20.000+0200},
  title = {Early Experience in Benchmarking Edge AI Processors with Object Detection Workloads.},
  url = {http://dblp.uni-trier.de/db/conf/bench/bench2019.html#HuiLL19},
  volume = 12093,
  series = {Bench '19},
  highlight = {Best Paper Award Nomination},
  year = 2019
}

@inproceedings{conf-nvmw-shankar19,
  author = {Shankar, Dipti and Lu, Xiaoyi and Panda, Dhabaleswar K.},
  Date-Added = {2019-02-22 19:42:43 +0000},
  booktitle = {Proceedings of Annual Non-Volatile Memories Workshop},
  crossref = {conf/nvmw/2019},
  keywords = {Non-Volatile Memory, In-Memory Datastore, Remote Direct Memory Persistence},
  title = {Accelerating NVRAM-aware In-Memory Datastore with Remote Direct Memory Persistence},
  series = {NVMW '19},
  note = {Poster Paper},
  year = {2019}
}

@inproceedings{conf-sc-hui19,
author = {Hui, Yujie and Lien, Jeffrey and Lu, Xiaoyi},
title = {Three-Dimensional Characterization on Edge AI Processors with Object Detection Workloads},
year = {2019},
publisher = {Association for Computing Machinery},
address = {New York, NY, USA},
booktitle = {Proceedings of the International Conference for High Performance Computing, Networking, Storage and Analysis},
keywords = {Edge Computing, AI, Object Detection},
location = {Denver, Colorado},
note = {Poster Paper},
series = {SC '19}
}

@ARTICLE{journals-debu-LuSP17,
  author={Chu, Ching-Hsiang and Lu, Xiaoyi and Awan, Ammar and Subramoni, Hari and Elton, Bracy and Panda, Dhabaleswar K.},
  journal={IEEE Transactions on Parallel and Distributed Systems}, 
  title={Exploiting Hardware Multicast and GPUDirect RDMA for Efficient Broadcast}, 
  year={2019},
  volume={30},
  number={3},
  pages={575-588},
  abstract={Broadcast is a widely used operation in many streaming and deep learning applications to disseminate large amounts of data on emerging heterogeneous High-Performance Computing (HPC) systems. However, traditional broadcast schemes do not fully utilize hardware features for Graphics Processing Unit (GPU)-based applications. In this paper, a model-oriented analysis is presented to identify performance bottlenecks of existing broadcast schemes on GPU clusters. Next, streaming-based broadcast schemes are proposed to exploit InfiniBand hardware multicast (IB-MCAST) and NVIDIA GPUDirect technology for efficient message transmission. The proposed designs are evaluated in the context of using Message Passing Interface (MPI) based benchmarks and applications. The experimental results indicate improved scalability and up to 82 percent reduction of latency compared to the state-of-the-art solutions in the benchmark-level evaluation. Furthermore, compared to the state-of-the-art, the proposed design yields stable higher throughput for a synthetic streaming workload, and 1.3x faster training time for a deep learning framework.},
  keywords={graphics processing units;learning (artificial intelligence);message passing;neural nets;parallel processing;performance evaluation;model-oriented analysis;GPU clusters;streaming-based broadcast schemes;InfiniBand hardware multicast;IB-MCAST;NVIDIA GPUDirect technology;streaming learning applications;message transmission;deep learning;high-performance computing;graphics processing unit-based applications;message passing interface;remote direct memory access technology;GPUDirect RDMA technology;HPC clusters;performance evaluation;Graphics processing units;Hardware;Analytical models;Machine learning;Clustering algorithms;Scalability;Bandwidth;Broadcast;deep learning;hardware multicast;GPU;GPUDirect RDMA;heterogeneous broadcast;streaming},
  doi={10.1109/TPDS.2018.2867222},
  ISSN={1558-2183},
  series={TPDS '19},
  month={March},}

@article{haseeb-thpc19,
	Abstract = {Over the last decade, technologies derived from convolutional neural networks (CNNs) called Deep Learning applications, have revolutionized fields as diverse as cancer detection, self-driving cars, virtual assistants, etc. However, many users of such applications are not experts in Machine Learning itself. Consequently, there is limited knowledge among the community to run such applications in an optimized manner. The performance question for Deep Learning applications has typically been addressed by employing bespoke hardware (e.g., GPUs) better suited for such compute-intensive operations. However, such a degree of performance is only accessibly at increasingly high financial costs leaving only big corporations and governments with resources sufficient enough to employ them at a large scale. As a result, an average user is only left with access to commodity clusters with, in many cases, only CPUs as the sole processing element. For such users to make effective use of resources at their disposal, concerted efforts are necessary to figure out optimal hardware and software configurations. This study is one such step in this direction as we use the Roofline model to perform a systematic analysis of representative CNN models and identify opportunities for black box and application-aware optimizations. Using the findings from our study, we are able to obtain up to 3.5{\$}{\$}{$\backslash$}times{\$}{\$}×speedup compared to vanilla TensorFlow with default configurations.},
	Author = {Javed, M. Haseeb and Ibrahim, Khaled Z. and Lu, Xiaoyi},
	Da = {2019/12/01},
	Date-Added = {2021-01-26 19:28:06 +0000},
	Date-Modified = {2021-01-26 19:28:52 +0000},
	Doi = {10.1007/s42514-019-00018-4},
	Id = {Javed2019},
	Isbn = {2524-4930},
	Journal = {CCF Transactions on High Performance Computing},
	Number = {3},
	Pages = {224--239},
	Title = {Performance analysis of deep learning workloads using roofline trajectories},
	Ty = {JOUR},
	Url = {https://doi.org/10.1007/s42514-019-00018-4},
	Volume = {1},
	Year = {2019},
	series = {THPC '19},
	note = {Invited Paper},
	Bdsk-Url-1 = {https://doi.org/10.1007/s42514-019-00018-4}}

@inproceedings{conf-iiswc-ShankarLP19,
  added-at = {2020-03-23T00:00:00.000+0100},
  author = {Shankar, Dipti and Lu, Xiaoyi and Panda, Dhabaleswar K. D. K.},
  biburl = {https://www.bibsonomy.org/bibtex/21fe2ee0c386a2946768b3f548a616ba9/dblp},
  booktitle = {Proceedings of IEEE International Symposium on Workload Characterization},
  crossref = {conf/iiswc/2019},
  ee = {https://doi.org/10.1109/IISWC47752.2019.9042069},
  interhash = {c42b7b3b8a7db2b799ab87e04014e9b2},
  intrahash = {1fe2ee0c386a2946768b3f548a616ba9},
  isbn = {978-1-7281-4045-2},
  keywords = {dblp},
  pages = {178-188},
  publisher = {IEEE},
  timestamp = {2020-03-24T11:40:24.000+0100},
  title = {SimdHT-Bench: Characterizing SIMD-Aware Hash Table Designs on Emerging CPU Architectures.},
  url = {http://dblp.uni-trier.de/db/conf/iiswc/iiswc2019.html#ShankarLP19},
  series = {IISWC '19},
  highlight = {Best Paper Award Nomination},
  year = 2019
}

@inproceedings{conf-sc-ShiL19,
author = {Shi, Haiyang and Lu, Xiaoyi},
title = {TriEC: Tripartite Graph Based Erasure Coding NIC Offload},
year = {2019},
isbn = {9781450362290},
publisher = {Association for Computing Machinery},
address = {New York, NY, USA},
url = {https://doi.org/10.1145/3295500.3356178},
doi = {10.1145/3295500.3356178},
abstract = {Erasure Coding (EC) NIC offload is a promising technology for designing next-generation distributed storage systems. However, this paper has identified three major limitations of current-generation EC NIC offload schemes on modern SmartNICs. Thus, this paper proposes a new EC NIC offload paradigm based on the tripartite graph model, namely TriEC. TriEC supports both encode-and-send and receive-and-decode operations efficiently. Through theorem-based proofs, co-designs with memcached (i.e., TriEC-Cache), and extensive experiments, we show that TriEC is correct and can deliver better performance than the state-of-the-art EC NIC offload schemes (i.e., BiEC). Benchmark evaluations demonstrate that TriEC outperforms BiEC by up to 1.82x and 2.33x for encoding and recovering, respectively. With extended YCSB workloads, TriEC reduces the average write latency by up to 23.2\% and the recovery time by up to 37.8\%. TriEC outperforms BiEC by 1.32x for a full-node recovery with 8 million records.},
booktitle = {Proceedings of the International Conference for High Performance Computing, Networking, Storage and Analysis},
articleno = {44},
numpages = {34},
keywords = {NIC offload, tripartite, bipartite, erasure coding},
location = {Denver, Colorado},
highlight = {Best Student Paper Finalist},
series = {SC '19}
}

@inproceedings{conf-hpdc-ShiLSP19,
  added-at = {2019-06-24T00:00:00.000+0200},
  author = {Shi, Haiyang and Lu, Xiaoyi and Shankar, Dipti and Panda, Dhabaleswar K.},
  biburl = {https://www.bibsonomy.org/bibtex/2d058cc4179255ed2e5236db637d5f1db/dblp},
  booktitle = {Proceedings of International ACM Symposium on High Performance and Distributed Computing},
  crossref = {conf/hpdc/2019},
  editor = {Weissman, Jon B. and Butt, Ali Raza and Smirni, Evgenia},
  ee = {https://doi.org/10.1145/3307681.3325406},
  interhash = {135a66dbfc9e45177209d2b865899c6a},
  intrahash = {d058cc4179255ed2e5236db637d5f1db},
  isbn = {978-1-4503-6670-0},
  keywords = {dblp},
  pages = {219-230},
  publisher = {ACM},
  timestamp = {2019-06-25T11:40:11.000+0200},
  title = {UMR-EC: A Unified and Multi-Rail Erasure Coding Library for High-Performance Distributed Storage Systems.},
  url = {http://dblp.uni-trier.de/db/conf/hpdc/hpdc2019.html#ShiLSP19},
  series = {HPDC '19},
  year = 2019
}

@inproceedings{conf-hipc-ShankarLP19,
  added-at = {2020-02-20T00:00:00.000+0100},
  author = {Shankar, Dipti and Lu, Xiaoyi and Panda, Dhabaleswar K.},
  biburl = {https://www.bibsonomy.org/bibtex/25d706fdfbff71fb7f02b813048fb9255/dblp},
  booktitle = {Proceedings of IEEE International Conference on High Performance Computing},
  crossref = {conf/hipc/2019},
  ee = {https://doi.org/10.1109/HiPC.2019.00040},
  interhash = {11642abcd32b35c9a7558de73f6172f2},
  intrahash = {5d706fdfbff71fb7f02b813048fb9255},
  isbn = {978-1-7281-4535-8},
  keywords = {dblp},
  pages = {257-266},
  publisher = {IEEE},
  timestamp = {2020-02-21T13:15:33.000+0100},
  title = {SCOR-KV: SIMD-Aware Client-Centric and Optimistic RDMA-Based Key-Value Store for Emerging CPU Architectures.},
  url = {http://dblp.uni-trier.de/db/conf/hipc/hipc2019.html#ShankarLP19},
  series = {HiPC '19},
  year = 2019
}

@inproceedings{conf-sc-shi19-poster,
author = {Shi, Haiyang and Lu, Xiaoyi},
title = {Designing High-Performance Erasure Coding Schemes for Next-Generation Storage Systems},
year = {2019},
publisher = {Association for Computing Machinery},
address = {New York, NY, USA},
booktitle = {Proceedings of the International Conference for High Performance Computing, Networking, Storage and Analysis},
keywords = {Erasure Coding, Storage Systems, SmartNIC, GPGPU},
location = {Denver, Colorado},
note = {ACM Student Research Competition Poster},
series = {SC '19}
}

@inproceedings{conf-ipps-0045LCP19,
  added-at = {2019-09-09T00:00:00.000+0200},
  author = {Zhang, Jie and Lu, Xiaoyi and Chu, Ching-Hsiang and Panda, Dhabaleswar K.},
  biburl = {https://www.bibsonomy.org/bibtex/270551382bbb333ff00d2836b54f5bf98/dblp},
  booktitle = {Proceedings of IEEE International Parallel and Distributed Processing Symposium},
  crossref = {conf/ipps/2019},
  ee = {https://doi.org/10.1109/IPDPS.2019.00034},
  interhash = {cfd466e5c213165b44d2b685122ef49e},
  intrahash = {70551382bbb333ff00d2836b54f5bf98},
  isbn = {978-1-7281-1246-6},
  keywords = {dblp},
  pages = {242-251},
  publisher = {IEEE},
  timestamp = {2019-10-17T20:34:23.000+0200},
  title = {C-GDR: High-Performance Container-Aware GPUDirect MPI Communication Schemes on RDMA Networks.},
  url = {http://dblp.uni-trier.de/db/conf/ipps/ipdps2019.html#0045LCP19},
  series = {IPDPS '19},
  year = 2019
}

@inproceedings{conf-hipc-AwanCSLP18,
  added-at = {2019-02-15T00:00:00.000+0100},
  author = {Awan, Ammar Ahmad and Chu, Ching-Hsiang and Subramoni, Hari and Lu, Xiaoyi and Panda, Dhabaleswar K.},
  biburl = {https://www.bibsonomy.org/bibtex/217d808edc8310a9ba11c233b3e78ef96/dblp},
  booktitle = {Proceedings of IEEE International Conference on High Performance Computing},
  crossref = {conf/hipc/2018},
  ee = {https://doi.org/10.1109/HiPC.2018.00024},
  interhash = {e8c319a09572e716db236ca636faf2c2},
  intrahash = {17d808edc8310a9ba11c233b3e78ef96},
  isbn = {978-1-5386-8386-6},
  keywords = {dblp},
  pages = {143-152},
  publisher = {IEEE},
  timestamp = {2019-10-17T16:01:05.000+0200},
  title = {OC-DNN: Exploiting Advanced Unified Memory Capabilities in CUDA 9 and Volta GPUs for Out-of-Core DNN Training.},
  url = {http://dblp.uni-trier.de/db/conf/hipc/hipc2018.html#AwanCSLP18},
  series = {HiPC '18},
  year = 2018
}

@article{journals-jzusc-PandaLS18,
	Abstract = {With the significant advancement in emerging processor, memory, and networking technologies, exascale systems will become available in the next few years (2020--2022). As the exascale systems begin to be deployed and used, there will be a continuous demand to run next-generation applications with finer granularity, finer time-steps, and increased data sizes. Based on historical trends, next-generation applications will require postexascale systems during 2025--2035. In this study, we focus on the networking and communication challenges for post-exascale systems. Firstly, we present an envisioned architecture for post-exascale systems. Secondly, the challenges are summarized from different perspectives: heterogeneous networking technologies, high-performance communication and synchronization protocols, integrated support with accelerators and field-programmable gate arrays, fault-tolerance and quality-of-service support, energy-aware communication schemes and protocols, softwaredefined networking, and scalable communication protocols with heterogeneous memory and storage. Thirdly, we present the challenges in designing efficient programming model support for high-performance computing, big data, and deep learning on these systems. Finally, we emphasize the critical need for co-designing runtime with upper layers on these systems to achieve the maximum performance and scalability.},
	Author = {Panda, Dhabaleswar and Lu, Xiaoyi and Subramoni, Hari},
	Da = {2018/10/01},
	Date-Added = {2021-02-03 03:45:44 +0000},
	Date-Modified = {2021-02-03 03:45:44 +0000},
	Doi = {10.1631/FITEE.1800631},
	Id = {Panda2018},
	Isbn = {2095-9230},
	Journal = {Frontiers of Information Technology \& Electronic Engineering},
	Number = {10},
	Pages = {1230--1235},
	Title = {Networking and communication challenges for post-exascale systems},
	Ty = {JOUR},
	Url = {https://doi.org/10.1631/FITEE.1800631},
	Volume = {19},
	Year = {2018},
	series = {FITEE '18},
	note = {Invited Vision Paper},
	Bdsk-Url-1 = {https://doi.org/10.1631/FITEE.1800631}}

@ARTICLE {journals-tmscs-LuSBJP18,
author = {Lu, Xiaoyi and Shi, Haiyang and Biswas, Rarjashi and Javed, M. Haseeb and Panda, Dhabaleswar K.},
journal = {IEEE Transactions on Multi-Scale Computing Systems},
title = {DLoBD: A Comprehensive Study of Deep Learning over Big Data Stacks on HPC Clusters},
year = {2018},
volume = {4},
number = {04},
issn = {2332-7766},
pages = {635-648},
keywords = {machine learning;big data;graphics processing units;training data;parallel processing;deep learning},
doi = {10.1109/TMSCS.2018.2845886},
publisher = {IEEE Computer Society},
address = {Los Alamitos, CA, USA},
series = {TMSCS '18},
month = {oct}
}

@inproceedings{conf-bench-JiangG0XZWLYLZF18,
  added-at = {2020-10-16T00:00:00.000+0200},
  author = {Jiang, Zihan and Gao, Wanling and Wang, Lei and Xiong, Xingwang and Zhang, Yuchen and Wen, Xu and Luo, Chunjie and Ye, Hainan and Lu, Xiaoyi and Zhang, Yunquan and Feng, Shengzhong and Li, Kenli and Xu, Weijia and Zhan, Jianfeng},
  biburl = {https://www.bibsonomy.org/bibtex/252053e143124a37d8ffb691807f936f4/dblp},
  booktitle = {Proceedings of International Symposium on Benchmarking, Measuring, and Optimizing},
  crossref = {conf/bench/2018},
  editor = {Zheng, Chen and Zhan, Jianfeng},
  ee = {https://doi.org/10.1007/978-3-030-32813-9_2},
  interhash = {86517bf9ea43796b1b0dbeb0f982fd02},
  intrahash = {52053e143124a37d8ffb691807f936f4},
  isbn = {978-3-030-32813-9},
  keywords = {dblp},
  pages = {10-22},
  publisher = {Springer},
  timestamp = {2020-10-17T11:35:55.000+0200},
  title = {HPC AI500: A Benchmark Suite for HPC AI Systems.},
  url = {http://dblp.uni-trier.de/db/conf/bench/bench2018.html#JiangG0XZWLYLZF18},
  volume = 11459,
  series = {Bench '18},
  year = 2018
}

@inproceedings{conf-bigdataconf-LuSSP18,
  added-at = {2019-01-27T00:00:00.000+0100},
  author = {Lu, Xiaoyi and Shankar, Dipti and Shi, Haiyang and Panda, Dhabaleswar K.},
  biburl = {https://www.bibsonomy.org/bibtex/2df3295e3dfa9d243e08350287ebe7810/dblp},
  booktitle = {Proceedings of IEEE International Conference on Big Data},
  crossref = {conf/bigdataconf/2018},
  editor = {Abe, Naoki and Liu, Huan and Pu, Calton and Hu, Xiaohua and Ahmed, Nesreen K. and Qiao, Mu and Song, Yang and Kossmann, Donald and Liu, Bing and Lee, Kisung and Tang, Jiliang and He, Jingrui and Saltz, Jeffrey S.},
  ee = {https://doi.org/10.1109/BigData.2018.8622615},
  interhash = {3a3a4feaf75a9b47b1f80c82d08091ce},
  intrahash = {df3295e3dfa9d243e08350287ebe7810},
  isbn = {978-1-5386-5035-6},
  keywords = {dblp},
  pages = {321-326},
  publisher = {IEEE},
  timestamp = {2020-03-20T11:49:51.000+0100},
  title = {Spark-uDAPL: Cost-Saving Big Data Analytics on Microsoft Azure Cloud with RDMA Networks*.},
  url = {http://dblp.uni-trier.de/db/conf/bigdataconf/bigdataconf2018.html#LuSSP18},
  series = {BigData '18},
  note = {Short Paper},
  year = 2018
}

@inproceedings{conf-bpoe-rajarshi,
  author = {Biswas, Rajarshi and Lu, Xiaoyi and Panda, Dhabaleswar K.},
  booktitle = {Proceedings of International Workshop on Big Data Benchmarks, Performance Optimization, and Emerging Hardware},
  crossref = {conf/bpoe/2018},
  title = {Designing a Micro-Benchmark Suite to Evaluate gRPC for TensorFlow: Early Experiences},
  url = {https://arxiv.org/abs/1804.01138},
  pdf = {https://arxiv.org/pdf/1804.01138.pdf},
  series = {BPOE '18},
  year = 2018
}

@inproceedings{conf-pvm-LiLSP18,
  added-at = {2018-11-21T00:00:00.000+0100},
  author = {Li, Mingzhe and Lu, Xiaoyi and Subramoni, Hari and Panda, Dhabaleswar K.},
  biburl = {https://www.bibsonomy.org/bibtex/20b9ef28e2c26443d6709cd917b709891/dblp},
  booktitle = {Proceedings of International Conference on EuroMPI},
  crossref = {conf/pvm/2018},
  ee = {https://doi.org/10.1145/3236367.3236371},
  interhash = {5200ed4573272fee26b3e4ef352eb17f},
  intrahash = {0b9ef28e2c26443d6709cd917b709891},
  keywords = {dblp},
  pages = {4:1-4:10},
  publisher = {ACM},
  timestamp = {2018-11-22T11:50:10.000+0100},
  title = {Multi-Threading and Lock-Free MPI RMA Based Graph Processing on KNL and POWER Architectures.},
  url = {http://dblp.uni-trier.de/db/conf/pvm/eurompi2018.html#LiLSP18},
  series = {EuroMPI '18},
  year = 2018
}

@inproceedings{conf-isc-awan18,
  added-at = {2018-07-23T00:00:00.000+0200},
  author = {Awan, Ammar Ahmad and Chu, Ching-Hsiang and Subramoni, Hari and Lu, Xiaoyi and Panda, Dhabaleswar K.},
  booktitle = {Proceedings of the International Supercomputing Conference},
  crossref = {conf/supercomputer/2018},
  keywords = {Unified Memory, Pascal, Volta, Out-of-Core DNN Training},
  publisher = {Springer},
  title = {Can Unified-Memory Support on Pascal and Volta GPUs enable Out-of-Core DNN Training?},
  pdf = {https://www.isc-hpc.com/agenda2018/conferences/isc_hpc/assets/2018/posters/post127.pdf},
  highlight = {Best Poster Paper Award},
  series = {ISC '18},
  note = {Poster Paper},
  year = 2018
}

@inproceedings{conf-bench-ZhangZLTLLWL18,
  added-at = {2019-10-22T00:00:00.000+0200},
  author = {Zhang, Qin and Zha, Li and Lin, Jian and Tu, Dandan and Li, Mingzhe and Liang, Fan and Wu, Ren and Lu, Xiaoyi},
  biburl = {https://www.bibsonomy.org/bibtex/24fd5e0876e08c50c88adb0dba1679d3a/dblp},
  booktitle = {Proceedings of International Symposium on Benchmarking, Measuring, and Optimizing},
  crossref = {conf/bench/2018},
  editor = {Zheng, Chen and Zhan, Jianfeng},
  ee = {https://doi.org/10.1007/978-3-030-32813-9_5},
  interhash = {ab0b2986c66edda962fde0db319dc496},
  intrahash = {4fd5e0876e08c50c88adb0dba1679d3a},
  isbn = {978-3-030-32813-9},
  keywords = {dblp},
  pages = {36-49},
  publisher = {Springer},
  timestamp = {2019-10-23T11:40:09.000+0200},
  title = {A Survey on Deep Learning Benchmarks: Do We Still Need New Ones?},
  url = {http://dblp.uni-trier.de/db/conf/bench/bench2018.html#ZhangZLTLLWL18},
  volume = 11459,
  series = {Bench '18},
  year = 2018
}

@article{journals-jpdc-Wasi-ur-RahmanI18,
title = "MR-Advisor: A comprehensive tuning, profiling, and prediction tool for MapReduce execution frameworks on HPC clusters",
journal = "Journal of Parallel and Distributed Computing",
volume = "120",
pages = "237 - 250",
year = "2018",
issn = "0743-7315",
doi = "https://doi.org/10.1016/j.jpdc.2017.11.004",
url = "http://www.sciencedirect.com/science/article/pii/S0743731517303027",
author = "Md. Wasi-ur- Rahman and Nusrat Sharmin Islam and Xiaoyi Lu and Dipti Shankar and Dhabaleswar K. (DK) Panda",
keywords = "MapReduce, HPC, Tuning, Prediction",
series = "JPDC '18",
abstract = "MapReduce is the most popular parallel computing framework for big data processing which allows massive scalability across distributed computing environment. Advanced RDMA-based design of Hadoop MapReduce has been proposed that alleviates the performance bottlenecks in default Hadoop MapReduce by leveraging the benefits from RDMA. On the other hand, data processing engine, Spark, provides fast execution of MapReduce applications through in-memory processing. Performance optimization for these contemporary big data processing frameworks on modern High-Performance Computing (HPC) systems is a formidable task because of the numerous configuration possibilities in each of them. In this paper, we propose MR-Advisor, a comprehensive tuning, profiling, and prediction tool for MapReduce. MR-Advisor is generalized to provide performance optimizations for Hadoop, Spark, and RDMA-enhanced Hadoop MapReduce designs over different file systems such as HDFS, Lustre, and Tachyon. Performance evaluations reveal that, with MR-Advisor’s suggested values, the job execution performance can be enhanced by a maximum of 58\% over the current best-practice values for user-level configuration parameters. To the best of our knowledge, this is the first tool that supports tuning and prediction for both Apache Hadoop and Spark, as well as the RDMA and Lustre-based advanced designs."
}

@inproceedings{conf-ucc-GugnaniLP18,
  added-at = {2019-02-15T00:00:00.000+0100},
  author = {Gugnani, Shashank and Lu, Xiaoyi and Panda, Dhabaleswar K.},
  biburl = {https://www.bibsonomy.org/bibtex/2deb73eb7d6de899432ac543f34be990f/dblp},
  booktitle = {Proceedings of IEEE/ACM International Conference on Utility and Cloud Computing},
  crossref = {conf/ucc/2018},
  editor = {Sill, Alan and Spillner, Josef},
  ee = {http://doi.ieeecomputersociety.org/10.1109/UCC.2018.00033},
  interhash = {399882fee2ec925cfd467e9ec73b0fb3},
  intrahash = {deb73eb7d6de899432ac543f34be990f},
  isbn = {978-1-5386-5504-7},
  keywords = {dblp},
  pages = {247-256},
  publisher = {IEEE Computer Society},
  timestamp = {2019-02-19T11:46:01.000+0100},
  title = {Analyzing, Modeling, and Provisioning QoS for NVMe SSDs.},
  url = {http://dblp.uni-trier.de/db/conf/ucc/ucc2018.html#GugnaniLP18},
  series = {UCC '18},
  year = 2018
}

@inproceedings{conf-nvmw-lu18,
  author = {Lu, Xiaoyi and Wasi-ur-Rahman, Md. and Islam, Nusrat and Panda, Dhabaleswar K.},
  Date-Added = {2018-02-22 19:42:43 +0000},
  booktitle = {Proceedings of Annual Non-Volatile Memories Workshop},
  crossref = {conf/nvmw/2018},
  keywords = {Non-Volatile Memory, RDMA, MapReduce, DAG},
  title = {Accelerating MapReduce and DAG Execution Frameworks with Non-Volatile Memory and RDMA},
  series = {NVMW '18},
  note = {Poster Paper},
  year = {2018}
}

@inproceedings{conf-bench-ShiLP18,
  added-at = {2019-10-22T00:00:00.000+0200},
  author = {Shi, Haiyang and Lu, Xiaoyi and Panda, Dhabaleswar K.},
  biburl = {https://www.bibsonomy.org/bibtex/220007ce33a36baf22f1ab92a0040622f/dblp},
  booktitle = {Proceedings of International Symposium on Benchmarking, Measuring, and Optimizing},
  crossref = {conf/bench/2018},
  editor = {Zheng, Chen and Zhan, Jianfeng},
  ee = {https://doi.org/10.1007/978-3-030-32813-9_18},
  interhash = {1103472dbf1d6e494219ed5fdf97f1f1},
  intrahash = {20007ce33a36baf22f1ab92a0040622f},
  isbn = {978-3-030-32813-9},
  keywords = {dblp},
  pages = {215-230},
  publisher = {Springer},
  timestamp = {2019-10-23T11:40:09.000+0200},
  title = {EC-Bench: Benchmarking Onload and Offload Erasure Coders on Modern Hardware Architectures.},
  url = {http://dblp.uni-trier.de/db/conf/bench/bench2018.html#ShiLP18},
  volume = 11459,
  series = {Bench '18},
  highlight = {Best Paper Award},
  year = 2018
}

@inproceedings{conf-hipc-BiswasLP18,
  added-at = {2019-02-15T00:00:00.000+0100},
  author = {Biswas, Rajarshi and Lu, Xiaoyi and Panda, Dhabaleswar K.},
  biburl = {https://www.bibsonomy.org/bibtex/2c27243aa3d97045723f542a6c3db3a6d/dblp},
  booktitle = {Proceedings of IEEE International Conference on High Performance Computing},
  crossref = {conf/hipc/2018},
  ee = {https://doi.org/10.1109/HiPC.2018.00010},
  interhash = {87aee73f13592e20bf8ae79e4c59e15e},
  intrahash = {c27243aa3d97045723f542a6c3db3a6d},
  isbn = {978-1-5386-8386-6},
  keywords = {dblp},
  pages = {2-11},
  publisher = {IEEE},
  timestamp = {2019-10-17T16:01:05.000+0200},
  title = {Accelerating TensorFlow with Adaptive RDMA-Based gRPC.},
  url = {http://dblp.uni-trier.de/db/conf/hipc/hipc2018.html#BiswasLP18},
  series = {HiPC '18},
  year = 2018
}

@inproceedings{conf-cloud-ShiLSP18,
  added-at = {2018-11-21T00:00:00.000+0100},
  author = {Shi, Haiyang and Lu, Xiaoyi and Shankar, Dipti and Panda, Dhabaleswar K.},
  biburl = {https://www.bibsonomy.org/bibtex/278ce655ad50d6f2571fb6acd45b55b15/dblp},
  booktitle = {Proceedings of ACM Symposium on Cloud Computing},
  crossref = {conf/cloud/2018},
  ee = {https://doi.org/10.1145/3267809.3275472},
  interhash = {adac90142b8ceb55ca56258ecd116b34},
  intrahash = {78ce655ad50d6f2571fb6acd45b55b15},
  isbn = {978-1-4503-6011-1},
  keywords = {dblp},
  pages = {530-531},
  publisher = {ACM},
  timestamp = {2018-11-22T12:00:12.000+0100},
  title = {High-Performance Multi-Rail Erasure Coding Library over Modern Data Center Architectures: Early Experiences.},
  url = {http://dblp.uni-trier.de/db/conf/cloud/socc2018.html#ShiLSP18},
  series = {SoCC '18},
  note = {Poster Paper},
  year = 2018
}

@inproceedings{conf-cluster-JavedLP18,
  added-at = {2018-12-23T00:00:00.000+0100},
  author = {Javed, M. Haseeb and Lu, Xiaoyi and Panda, Dhabaleswar K.},
  biburl = {https://www.bibsonomy.org/bibtex/266411be31238baadfd2ed154951b0ae8/dblp},
  booktitle = {Proceedings of IEEE International Conference on Cluster Computing},
  crossref = {conf/cluster/2018},
  ee = {http://doi.ieeecomputersociety.org/10.1109/CLUSTER.2018.00040},
  interhash = {55d49f4d21110d8c7d48dc02d3939202},
  intrahash = {66411be31238baadfd2ed154951b0ae8},
  isbn = {978-1-5386-8319-4},
  keywords = {dblp},
  pages = {223-233},
  publisher = {IEEE Computer Society},
  timestamp = {2018-12-25T11:40:58.000+0100},
  title = {Cutting the Tail: Designing High Performance Message Brokers to Reduce Tail Latencies in Stream Processing.},
  url = {http://dblp.uni-trier.de/db/conf/cluster/cluster2018.html#JavedLP18},
  series = {CLUSTER '18},
  year = 2018
}

@inproceedings{conf-hoti-LuSJBP17,
  added-at = {2017-10-27T00:00:00.000+0200},
  author = {Lu, Xiaoyi and Shi, Haiyang and Javed, M. Haseeb and Biswas, Rajarshi and Panda, Dhabaleswar K.},
  biburl = {https://www.bibsonomy.org/bibtex/213a49f47443e2dd2a537e6de4c4af16e/dblp},
  booktitle = {Proceedings of Annual Symposium on High-Performance Interconnects},
  crossref = {conf/hoti/2017},
  ee = {http://doi.ieeecomputersociety.org/10.1109/HOTI.2017.24},
  interhash = {31903f8a5fdca3db2070815403d84b64},
  intrahash = {13a49f47443e2dd2a537e6de4c4af16e},
  isbn = {978-1-5386-1013-8},
  keywords = {dblp},
  pages = {87-94},
  publisher = {IEEE Computer Society},
  timestamp = {2017-12-07T22:37:35.000+0100},
  title = {Characterizing Deep Learning over Big Data (DLoBD) Stacks on RDMA-Capable Networks.},
  url = {http://dblp.uni-trier.de/db/conf/hoti/hoti2017.html#LuSJBP17},
  series = {HotI '17},
  year = 2017
}

@inproceedings{conf-icdcs-ShankarLP17,
  added-at = {2018-04-11T00:00:00.000+0200},
  author = {Shankar, Dipti and Lu, Xiaoyi and Panda, Dhabaleswar K.},
  biburl = {https://www.bibsonomy.org/bibtex/2ffd5e0e807b8860727aa791416118b56/dblp},
  booktitle = {Proceedings of IEEE International Conference on Distributed Computing Systems},
  crossref = {conf/icdcs/2017},
  editor = {Lee, Kisung and Liu, Ling},
  ee = {http://doi.ieeecomputersociety.org/10.1109/ICDCS.2017.224},
  interhash = {98c6d49fea5aa19383e46d20638aca3d},
  intrahash = {ffd5e0e807b8860727aa791416118b56},
  isbn = {978-1-5386-1792-2},
  keywords = {dblp},
  pages = {527-537},
  publisher = {IEEE Computer Society},
  timestamp = {2018-04-12T11:45:53.000+0200},
  title = {High-Performance and Resilient Key-Value Store with Online Erasure Coding for Big Data Workloads.},
  url = {http://dblp.uni-trier.de/db/conf/icdcs/icdcs2017.html#ShankarLP17},
  series = {ICDCS '17},
  year = 2017
}

@ARTICLE{journals-tpds-Wasi-ur-RahmanI17,
  author={Wasi-ur-Rahman, Md. and Islam, Nusrat Sharmin and Lu, Xiaoyi and Panda, Dhabaleswar K.},
  journal={IEEE Transactions on Parallel and Distributed Systems}, 
  title={A Comprehensive Study of MapReduce Over Lustre for Intermediate Data Placement and Shuffle Strategies on HPC Clusters}, 
  year={2017},
  volume={28},
  number={3},
  pages={633-646},
  abstract={With high performance interconnects and parallel file systems, running MapReduce over modern High Performance Computing (HPC) clusters has attracted much attention due to its uniqueness of solving data analytics problems with a combination of Big Data and HPC technologies. Since the MapReduce architecture relies heavily on the availability of local storage media, the Lustrebased global storage in HPC clusters poses many new opportunities and challenges. In this paper, we perform a comprehensive study on different MapReduce over Lustre deployments and propose a novel high-performance design of YARN MapReduce on HPC clusters by utilizing Lustre as the additional storage provider for intermediate data. With a deployment architecture where both local disks and Lustre are utilized for intermediate data storage, we propose a novel priority directory selection scheme through which RDMAenhanced MapReduce can choose the best intermediate storage during runtime by on-line profiling. Our results indicate that, we can achieve 44 percent performance benefit for shuffle-intensive workloads in leadership-class HPC systems. Our priority directory selection scheme can improve the job execution time by 63 percent over default MapReduce while executing multiple concurrent jobs. To the best of our knowledge, this is the first such comprehensive study for YARN MapReduce with Lustre and RDMA.},
  keywords={Big Data;distributed databases;network operating systems;parallel architectures;resource allocation;storage management;storage media;workstation clusters;intermediate data placement;HPC clusters;high performance interconnects;parallel file systems;high performance computing clusters;data analytics;Big Data;HPC technologies;local storage media;Lustre-based global storage;MapReduce over Lustre deployments;high-performance YARN MapReduce design;storage provider;intermediate data storage;priority directory selection;RDMA-enhanced MapReduce;shuffle-intensive workloads;leadership-class HPC systems;job execution;Computer architecture;Servers;High performance computing;Data analysis;Big data;Memory;Big data;high performance computing;RDMA;MapReduce;lustre},
  doi={10.1109/TPDS.2016.2591947},
  ISSN={1558-2183},
  series = {TPDS '17},
  month={March},}

@inproceedings{conf-bigdataconf-LuSSP17,
  added-at = {2019-02-20T00:00:00.000+0100},
  author = {Lu, Xiaoyi and Shi, Haiyang and Shankar, Dipti and Panda, Dhabaleswar K.},
  biburl = {https://www.bibsonomy.org/bibtex/2c2b4ada9aaba9c32a123ead1ae899258/dblp},
  booktitle = {Proceedings of IEEE International Conference on Big Data},
  crossref = {conf/bigdataconf/2017},
  editor = {Nie, Jian-Yun and Obradovic, Zoran and Suzumura, Toyotaro and Ghosh, Rumi and Nambiar, Raghunath and Wang, Chonggang and Zang, Hui and Baeza-Yates, Ricardo and Hu, Xiaohua and Kepner, Jeremy and Cuzzocrea, Alfredo and Tang, Jian and Toyoda, Masashi},
  ee = {http://doi.ieeecomputersociety.org/10.1109/BigData.2017.8257929},
  interhash = {cf28ebee4ca073c6b96e58cf3b152bcf},
  intrahash = {c2b4ada9aaba9c32a123ead1ae899258},
  isbn = {978-1-5386-2715-0},
  keywords = {dblp},
  pages = {213-222},
  publisher = {IEEE Computer Society},
  timestamp = {2019-09-24T11:44:56.000+0200},
  title = {Performance characterization and acceleration of big data workloads on OpenPOWER system.},
  url = {http://dblp.uni-trier.de/db/conf/bigdataconf/bigdataconf2017.html#LuSSP17},
  series = {BigData '17},
  year = 2017
}

@inproceedings{conf-hipc-GugnaniLPCP17,
  added-at = {2018-02-19T00:00:00.000+0100},
  author = {Gugnani, Shashank and Lu, Xiaoyi and Pestilli, Franco and Caiafa, Cesar F. and Panda, Dhabaleswar K.},
  biburl = {https://www.bibsonomy.org/bibtex/26475a8cefb868ccf66d6c5a5369c1c97/dblp},
  booktitle = {Proceedings of IEEE International Conference on High Performance Computing},
  crossref = {conf/hipc/2017},
  ee = {http://doi.ieeecomputersociety.org/10.1109/HiPC.2017.00033},
  interhash = {0c01a8b3e2244cc18e7ffd3257bfeb9e},
  intrahash = {6475a8cefb868ccf66d6c5a5369c1c97},
  isbn = {978-1-5386-2293-3},
  keywords = {dblp},
  pages = {213-222},
  publisher = {IEEE Computer Society},
  timestamp = {2019-10-17T16:00:49.000+0200},
  title = {MPI-LiFE: Designing High-Performance Linear Fascicle Evaluation of Brain Connectome with MPI.},
  url = {http://dblp.uni-trier.de/db/conf/hipc/hipc2017.html#GugnaniLPCP17},
  series = {HiPC '17},
  year = 2017
}

@inproceedings{conf-bdc-JavedLP17,
  added-at = {2018-11-06T00:00:00.000+0100},
  author = {Javed, M. Haseeb and Lu, Xiaoyi and Panda, Dhabaleswar K.},
  biburl = {https://www.bibsonomy.org/bibtex/2822966363c0bc004a37379b410d514d4/dblp},
  booktitle = {Proceedings of IEEE/ACM International Conference on Big Data Computing, Applications, and Technologies},
  crossref = {conf/bdc/2017},
  editor = {Anjum, Ashiq and Sill, Alan and Zhao, Xinghui and Farid, Mohsen M. and Pallickara, Shrideep and Cao, Jiannong},
  ee = {https://doi.org/10.1145/3148055.3148068},
  interhash = {fa924bec9b77bd7044b10b0a662f78ca},
  intrahash = {822966363c0bc004a37379b410d514d4},
  isbn = {978-1-4503-5549-0},
  keywords = {dblp},
  pages = {1-10},
  publisher = {ACM},
  timestamp = {2018-11-07T12:49:18.000+0100},
  title = {Characterization of Big Data Stream Processing Pipeline: A Case Study using Flink and Kafka.},
  url = {http://dblp.uni-trier.de/db/conf/bdc/bdcat2017.html#JavedLP17},
  series = {BDCAT '17},
  year = 2017
}

@inproceedings{conf-sc-Bayatpour0SLP17,
  added-at = {2018-11-06T00:00:00.000+0100},
  author = {Bayatpour, Mohammadreza and Chakraborty, Sourav and Subramoni, Hari and Lu, Xiaoyi and Panda, Dhabaleswar K.},
  biburl = {https://www.bibsonomy.org/bibtex/2e1c5377eae366b46bea3d4e169d668c0/dblp},
  booktitle = {Proceedings of International Conference for High Performance Computing, Networking, Storage and Analysis},
  crossref = {conf/sc/2017},
  editor = {Mohr, Bernd and Raghavan, Padma},
  ee = {https://doi.org/10.1145/3126908.3126954},
  interhash = {1e3eeb6e8d005488d2d7ea2d0eb680a5},
  intrahash = {e1c5377eae366b46bea3d4e169d668c0},
  isbn = {978-1-4503-5114-0},
  keywords = {dblp},
  pages = {64:1-64:11},
  publisher = {ACM},
  timestamp = {2019-09-26T12:39:54.000+0200},
  title = {Scalable reduction collectives with data partitioning-based multi-leader design.},
  url = {http://dblp.uni-trier.de/db/conf/sc/sc2017.html#Bayatpour0SLP17},
  series = {SC '17},
  year = 2017
}

@inproceedings{conf-vee-ZhangLP17,
  added-at = {2018-11-06T00:00:00.000+0100},
  author = {Zhang, Jie and Lu, Xiaoyi and Panda, Dhabaleswar K.},
  biburl = {https://www.bibsonomy.org/bibtex/2707d5d91a0dcd5efe196edfcd61320c1/dblp},
  booktitle = {Proceedings of ACM SIGPLAN/SIGOPS International Conference on Virtual Execution Environments},
  crossref = {conf/vee/2017},
  ee = {https://doi.org/10.1145/3050748.3050765},
  interhash = {c4876cfc6618ab18074e126d4067b5ce},
  intrahash = {707d5d91a0dcd5efe196edfcd61320c1},
  isbn = {978-1-4503-4948-2},
  keywords = {dblp},
  pages = {187-200},
  publisher = {ACM},
  timestamp = {2018-11-07T13:46:22.000+0100},
  title = {Designing Locality and NUMA Aware MPI Runtime for Nested Virtualization based HPC Cloud with SR-IOV Enabled InfiniBand.},
  url = {http://dblp.uni-trier.de/db/conf/vee/vee2017.html#ZhangLP17},
  series = {VEE '17},
  year = 2017
}

@inproceedings{conf-cluster-SubramoniLP17,
  added-at = {2018-02-19T00:00:00.000+0100},
  author = {Subramoni, Hari and Lu, Xiaoyi and Panda, Dhabaleswar K.},
  biburl = {https://www.bibsonomy.org/bibtex/29354bff1e63f65fa5685f86f55d4aade/dblp},
  booktitle = {Proceedings of IEEE International Conference on Cluster Computing},
  crossref = {conf/cluster/2017},
  ee = {http://doi.ieeecomputersociety.org/10.1109/CLUSTER.2017.78},
  interhash = {69a86a315131e6423ea85fef4585de13},
  intrahash = {9354bff1e63f65fa5685f86f55d4aade},
  isbn = {978-1-5386-2326-8},
  keywords = {dblp},
  pages = {354-358},
  publisher = {IEEE Computer Society},
  timestamp = {2018-02-21T11:43:46.000+0100},
  title = {A Scalable Network-Based Performance Analysis Tool for MPI on Large-Scale HPC Systems.},
  url = {http://dblp.uni-trier.de/db/conf/cluster/cluster2017.html#SubramoniLP17},
  series = {CLUSTER '17},
  note = {Short Paper},
  year = 2017
}

@inproceedings{conf-bigdataconf-Wasi-ur-RahmanI17,
  added-at = {2019-02-20T00:00:00.000+0100},
  author = {ur Rahman, Md. Wasi and Islam, Nusrat Sharmin and Lu, Xiaoyi and Panda, Dhabaleswar K.},
  biburl = {https://www.bibsonomy.org/bibtex/2381315c4108c4866c953e4384d9ede0d/dblp},
  booktitle = {Proceedings of IEEE International Conference on Big Data},
  crossref = {conf/bigdataconf/2017},
  editor = {Nie, Jian-Yun and Obradovic, Zoran and Suzumura, Toyotaro and Ghosh, Rumi and Nambiar, Raghunath and Wang, Chonggang and Zang, Hui and Baeza-Yates, Ricardo and Hu, Xiaohua and Kepner, Jeremy and Cuzzocrea, Alfredo and Tang, Jian and Toyoda, Masashi},
  ee = {http://doi.ieeecomputersociety.org/10.1109/BigData.2017.8257947},
  interhash = {275cbe7d990176256890daef49a900e5},
  intrahash = {381315c4108c4866c953e4384d9ede0d},
  isbn = {978-1-5386-2715-0},
  keywords = {dblp},
  pages = {369-374},
  publisher = {IEEE Computer Society},
  timestamp = {2019-09-24T11:44:56.000+0200},
  title = {NVMD: Non-volatile memory assisted design for accelerating MapReduce and DAG execution frameworks on HPC systems.},
  url = {http://dblp.uni-trier.de/db/conf/bigdataconf/bigdataconf2017.html#Wasi-ur-RahmanI17},
  series = {BigData '17},
  note = {Short Paper},
  year = 2017
}

@inproceedings{conf-nvmw-lu17,
  author = {Lu, Xiaoyi and Islam, Nusrat and Wasi-ur-Rahman, Md. and Panda, Dhabaleswar K.},
  Date-Added = {2017-02-22 19:42:43 +0000},
  booktitle = {Proceedings of Annual Non-Volatile Memories Workshop},
  crossref = {conf/nvmw/2017},
  keywords = {Non-Volatile Memory, RDMA, Big Data},
  title = {NRCIO: NVM-aware RDMA-based Communication and I/O Schemes for Big Data Analytics},
  series = {NVMW '17},
  note = {Position Paper},
  year = {2017}
}

@incollection{books-sp-17-Lu0P17,
  added-at = {2019-04-03T00:00:00.000+0200},
  author = {Lu, Xiaoyi and Zhang, Jie and Panda, Dhabaleswar K.},
  biburl = {https://www.bibsonomy.org/bibtex/209c64b6e7d73330583570e18ad0b7ad0/dblp},
  booktitle = {Research Advances in Cloud Computing},
  crossref = {books/sp/CSB2017},
  editor = {Chaudhary, Sanjay and Somani, Gaurav and Buyya, Rajkumar},
  ee = {https://doi.org/10.1007/978-981-10-5026-8_6},
  interhash = {c711bf152f733f286c36a8bf2ecc8672},
  intrahash = {09c64b6e7d73330583570e18ad0b7ad0},
  isbn = {978-981-10-5026-8},
  keywords = {dblp},
  pages = {115-140},
  publisher = {Springer},
  timestamp = {2019-05-30T11:48:04.000+0200},
  title = {Building Efficient HPC Cloud with SR-IOV-Enabled InfiniBand: The MVAPICH2 Approach.},
  url = {http://dblp.uni-trier.de/db/books/collections/CSB2017.html#Lu0P17},
  series = {Chapter '17},
  year = 2017
}

@inproceedings{conf-ucc-ZhangLP17,
  added-at = {2018-11-06T00:00:00.000+0100},
  author = {Zhang, Jie and Lu, Xiaoyi and Panda, Dhabaleswar K.},
  biburl = {https://www.bibsonomy.org/bibtex/20c5eae7b753ccc89cfe2de444d467f6f/dblp},
  booktitle = {Proceedings of IEEE/ACM International Conference on Utility and Cloud Computing},
  crossref = {conf/ucc/2017},
  editor = {Anjum, Ashiq and Sill, Alan and Fox, Geoffrey C. and Chen, Yong},
  ee = {https://doi.org/10.1145/3147213.3147231},
  interhash = {e0724c9854b4e433e861bf55e8f4d3db},
  intrahash = {0c5eae7b753ccc89cfe2de444d467f6f},
  isbn = {978-1-4503-5149-2},
  keywords = {dblp},
  pages = {151-160},
  publisher = {ACM},
  timestamp = {2018-11-07T13:40:43.000+0100},
  title = {Is Singularity-based Container Technology Ready for Running MPI Applications on HPC Clouds?},
  url = {http://dblp.uni-trier.de/db/conf/ucc/ucc2017.html#ZhangLP17},
  series = {UCC '17},
  highlight = {Best Student Paper Award},
  year = 2017
}

@inproceedings{conf-bigdataconf-GugnaniLQZP17,
  added-at = {2019-02-20T00:00:00.000+0100},
  author = {Gugnani, Shashank and Lu, Xiaoyi and Qi, Houliang and Zha, Li and Panda, Dhabaleswar K.},
  biburl = {https://www.bibsonomy.org/bibtex/2b74e5ce87ae07349fdd666cef2559728/dblp},
  booktitle = {Proceedings of IEEE International Conference on Big Data},
  crossref = {conf/bigdataconf/2017},
  editor = {Nie, Jian-Yun and Obradovic, Zoran and Suzumura, Toyotaro and Ghosh, Rumi and Nambiar, Raghunath and Wang, Chonggang and Zang, Hui and Baeza-Yates, Ricardo and Hu, Xiaohua and Kepner, Jeremy and Cuzzocrea, Alfredo and Tang, Jian and Toyoda, Masashi},
  ee = {http://doi.ieeecomputersociety.org/10.1109/BigData.2017.8257925},
  interhash = {feb10c50cc4194e162088f879531dfaf},
  intrahash = {b74e5ce87ae07349fdd666cef2559728},
  isbn = {978-1-5386-2715-0},
  keywords = {dblp},
  pages = {173-182},
  publisher = {IEEE Computer Society},
  timestamp = {2019-09-24T11:44:56.000+0200},
  title = {Characterizing and accelerating indexing techniques on distributed ordered tables.},
  url = {http://dblp.uni-trier.de/db/conf/bigdataconf/bigdataconf2017.html#GugnaniLQZP17},
  series = {BigData '17},
  year = 2017
}

@inproceedings{conf-icpp-ChuLASHEP17,
  added-at = {2017-09-12T00:00:00.000+0200},
  author = {Chu, Ching-Hsiang and Lu, Xiaoyi and Awan, Ammar Ahmad and Subramoni, Hari and Hashmi, Jahanzeb Maqbool and Elton, Bracy and Panda, Dhabaleswar K.},
  biburl = {https://www.bibsonomy.org/bibtex/2cefcaa45a20c67e0047302582017d24e/dblp},
  booktitle = {Proceedings of International Conference on Parallel Processing},
  crossref = {conf/icpp/2017},
  ee = {http://doi.ieeecomputersociety.org/10.1109/ICPP.2017.25},
  interhash = {a5ee04948fc3409affa91644baac9bc4},
  intrahash = {cefcaa45a20c67e0047302582017d24e},
  isbn = {978-1-5386-1042-8},
  keywords = {dblp},
  pages = {161-170},
  publisher = {IEEE Computer Society},
  timestamp = {2017-09-13T11:39:19.000+0200},
  title = {Efficient and Scalable Multi-Source Streaming Broadcast on GPU Clusters for Deep Learning.},
  url = {http://dblp.uni-trier.de/db/conf/icpp/icpp2017.html#ChuLASHEP17},
  series = {ICPP '17},
  year = 2017
}

@article{journals-tcde-LuSP17,
  added-at = {2020-03-10T00:00:00.000+0100},
  author = {Lu, Xiaoyi and Shankar, Dipti and Panda, Dhabaleswar K.},
  biburl = {https://www.bibsonomy.org/bibtex/2f34f6a1ad6702e0e89c4623a73bbae55/dblp},
  ee = {http://sites.computer.org/debull/A17mar/p50.pdf},
  interhash = {2be49a99d08c82b6198361b0c63605ef},
  intrahash = {f34f6a1ad6702e0e89c4623a73bbae55},
  journal = {IEEE Data Eng. Bull., Bulletin of the Technical Committee on Data Engineering},
  keywords = {dblp},
  number = 1,
  pages = {50-61},
  timestamp = {2020-03-11T11:39:57.000+0100},
  title = {Scalable and Distributed Key-Value Store-based Data Management Using RDMA-Memcached.},
  url = {http://dblp.uni-trier.de/db/journals/debu/debu40.html#LuSP17},
  volume = 40,
  series = {TCDE '17},
  note = {Invited Paper},
  year = 2017
}

@inproceedings{conf-ccgrid-GugnaniLP17,
  added-at = {2019-02-05T00:00:00.000+0100},
  author = {Gugnani, Shashank and Lu, Xiaoyi and Panda, Dhabaleswar K.},
  biburl = {https://www.bibsonomy.org/bibtex/2e8bd0869c082610c1af994f3b46c15ec/dblp},
  booktitle = {Proceedings of IEEE/ACM International Symposium on Cluster, Cloud and Grid Computing},
  crossref = {conf/ccgrid/2017},
  ee = {http://dl.acm.org/citation.cfm?id=3101145},
  interhash = {f03b5d9e6adc4b428db6f3a1824b6eb6},
  intrahash = {e8bd0869c082610c1af994f3b46c15ec},
  isbn = {978-1-5090-6610-0},
  keywords = {dblp},
  pages = {238-247},
  publisher = {IEEE Computer Society / ACM},
  timestamp = {2019-02-06T11:41:28.000+0100},
  title = {Swift-X: Accelerating OpenStack Swift with RDMA for Building an Efficient HPC Cloud.},
  url = {http://dblp.uni-trier.de/db/conf/ccgrid/ccgrid2017.html#GugnaniLP17},
  series = {CCGrid '17},
  year = 2017
}

@inproceedings{conf-ucc-PandaL17,
  added-at = {2018-11-06T00:00:00.000+0100},
  author = {Panda, Dhabaleswar K. and Lu, Xiaoyi},
  biburl = {https://www.bibsonomy.org/bibtex/2f836d1b3c517168465ae83f7acc446f2/dblp},
  booktitle = {Proceedings of IEEE/ACM International Conference on Utility and Cloud Computing},
  crossref = {conf/ucc/2017},
  editor = {Anjum, Ashiq and Sill, Alan and Fox, Geoffrey C. and Chen, Yong},
  ee = {https://doi.org/10.1145/3147213.3149455},
  interhash = {e457fd9b324278e70433de3cde4d62ae},
  intrahash = {f836d1b3c517168465ae83f7acc446f2},
  isbn = {978-1-4503-5149-2},
  keywords = {dblp},
  pages = {189-190},
  publisher = {ACM},
  timestamp = {2018-11-07T13:40:43.000+0100},
  title = {HPC Meets Cloud: Building Efficient Clouds for HPC, Big Data, and Deep Learning Middleware and Applications.},
  url = {http://dblp.uni-trier.de/db/conf/ucc/ucc2017.html#PandaL17},
  series = {UCC '17},
  year = 2017
}

@inproceedings{conf-hipc-LiLSP17,
  added-at = {2018-02-19T00:00:00.000+0100},
  author = {Li, Mingzhe and Lu, Xiaoyi and Subramoni, Hari and Panda, Dhabaleswar K.},
  biburl = {https://www.bibsonomy.org/bibtex/213393e0a324c2429862f132103c5f115/dblp},
  booktitle = {Proceedings of IEEE International Conference on High Performance Computing},
  crossref = {conf/hipc/2017},
  ee = {http://doi.ieeecomputersociety.org/10.1109/HiPC.2017.00017},
  interhash = {9d7bb5e8571ce872352214d1289028cd},
  intrahash = {13393e0a324c2429862f132103c5f115},
  isbn = {978-1-5386-2293-3},
  keywords = {dblp},
  pages = {62-71},
  publisher = {IEEE Computer Society},
  timestamp = {2019-10-17T16:00:49.000+0200},
  title = {Designing Registration Caching Free High-Performance MPI Library with Implicit On-Demand Paging (ODP) of InfiniBand.},
  url = {http://dblp.uni-trier.de/db/conf/hipc/hipc2017.html#LiLSP17},
  series = {HiPC '17},
  year = 2017
}

@inproceedings{conf-ipps-ZhangLP17,
  added-at = {2019-02-12T00:00:00.000+0100},
  author = {Zhang, Jie and Lu, Xiaoyi and Panda, Dhabaleswar K.},
  biburl = {https://www.bibsonomy.org/bibtex/2dcd8774d9b5f9511014355b216eb1129/dblp},
  booktitle = {Proceedings of IEEE International Parallel and Distributed Processing Symposium},
  crossref = {conf/ipps/2017},
  ee = {http://doi.ieeecomputersociety.org/10.1109/IPDPS.2017.43},
  interhash = {ed7753f38c6b07a75e7ce9118bce9ff3},
  intrahash = {dcd8774d9b5f9511014355b216eb1129},
  isbn = {978-1-5386-3914-6},
  keywords = {dblp},
  pages = {143-152},
  publisher = {IEEE Computer Society},
  timestamp = {2019-10-17T20:36:23.000+0200},
  title = {High-Performance Virtual Machine Migration Framework for MPI Applications on SR-IOV Enabled InfiniBand Clusters.},
  url = {http://dblp.uni-trier.de/db/conf/ipps/ipdps2017.html#ZhangLP17},
  series = {IPDPS '17},
  year = 2017
}

@inproceedings{conf-bdc-GugnaniLP16,
  added-at = {2018-11-06T00:00:00.000+0100},
  author = {Gugnani, Shashank and Lu, Xiaoyi and Panda, Dhabaleswar K.},
  biburl = {https://www.bibsonomy.org/bibtex/26645f5573a7d26c98c2a17c29e847c00/dblp},
  booktitle = {Proceedings of IEEE/ACM International Conference on Big Data Computing, Applications, and Technologies},
  crossref = {conf/bdc/2016},
  editor = {Anjum, Ashiq and Zhao, Xinghui},
  ee = {https://doi.org/10.1145/3006299.3006313},
  interhash = {bfc14acb8e51f6c145bb47feb624322e},
  intrahash = {6645f5573a7d26c98c2a17c29e847c00},
  isbn = {978-1-4503-4617-7},
  keywords = {dblp},
  pages = {36-45},
  publisher = {ACM},
  timestamp = {2019-02-06T11:37:22.000+0100},
  title = {Performance characterization of hadoop workloads on SR-IOV-enabled virtualized InfiniBand clusters.},
  url = {http://dblp.uni-trier.de/db/conf/bdc/bdcat2016.html#GugnaniLP16},
  series = {BDCAT '16},
  year = 2016
}

@inproceedings{conf-sbac-pad-Wasi-ur-RahmanI16,
  added-at = {2017-01-08T00:00:00.000+0100},
  author = {ur Rahman, Md. Wasi and Islam, Nusrat Sharmin and Lu, Xiaoyi and Shankar, Dipti and Panda, Dhabaleswar K.},
  biburl = {https://www.bibsonomy.org/bibtex/2d517803ac7a422525e997e8e5b430f66/dblp},
  booktitle = {Proceedings of International Symposium on Computer Architecture and High Performance Computing},
  crossref = {conf/sbac-pad/2016},
  ee = {http://doi.ieeecomputersociety.org/10.1109/SBAC-PAD.2016.33},
  interhash = {b5b3d869594bff6aa45d45066b79173b},
  intrahash = {d517803ac7a422525e997e8e5b430f66},
  isbn = {978-1-5090-6108-2},
  keywords = {dblp},
  pages = {198-205},
  publisher = {IEEE Computer Society},
  timestamp = {2017-01-10T11:36:45.000+0100},
  title = {MR-Advisor: A Comprehensive Tuning Tool for Advising HPC Users to Accelerate MapReduce Applications on Supercomputers.},
  url = {http://dblp.uni-trier.de/db/conf/sbac-pad/sbac-pad2016.html#Wasi-ur-RahmanI16},
  series = {SBAC-PAD '16},
  year = 2016
}

@inproceedings{conf-bigdataconf-LuSGP16,
  added-at = {2017-05-26T00:00:00.000+0200},
  author = {Lu, Xiaoyi and Shankar, Dipti and Gugnani, Shashank and Panda, Dhabaleswar K.},
  biburl = {https://www.bibsonomy.org/bibtex/293cfae12b194a4235c78b3d6d25c3cee/dblp},
  booktitle = {Proceedings of IEEE International Conference on Big Data},
  crossref = {conf/bigdataconf/2016},
  editor = {Joshi, James and Karypis, George and Liu, Ling and Hu, Xiaohua and Ak, Ronay and Xia, Yinglong and Xu, Weijia and Sato, Aki-Hiro and Rachuri, Sudarsan and Ungar, Lyle H. and Yu, Philip S. and Govindaraju, Rama and Suzumura, Toyotaro},
  ee = {https://doi.org/10.1109/BigData.2016.7840611},
  interhash = {e2bbfbe8f37033f34238d6556fdf5d1a},
  intrahash = {93cfae12b194a4235c78b3d6d25c3cee},
  isbn = {978-1-4673-9005-7},
  keywords = {dblp},
  pages = {253-262},
  publisher = {IEEE Computer Society},
  timestamp = {2019-02-21T11:45:26.000+0100},
  title = {High-performance design of apache spark with RDMA and its benefits on various workloads.},
  url = {http://dblp.uni-trier.de/db/conf/bigdataconf/bigdataconf2016.html#LuSGP16},
  series = {BigData '16},
  year = 2016
}

@inproceedings{conf-ipps-ShankarLIWP16,
  added-at = {2017-05-24T00:00:00.000+0200},
  author = {Shankar, Dipti and Lu, Xiaoyi and Islam, Nusrat S. and ur Rahman, Md. Wasi and Panda, Dhabaleswar K.},
  biburl = {https://www.bibsonomy.org/bibtex/2af1bd15d5a3e62e103ecb0a689545d3f/dblp},
  booktitle = {Proceedings of IEEE International Parallel and Distributed Processing Symposium},
  crossref = {conf/ipps/2016},
  ee = {http://doi.ieeecomputersociety.org/10.1109/IPDPS.2016.112},
  interhash = {ef70734acff5bff5ea65101f360069df},
  intrahash = {af1bd15d5a3e62e103ecb0a689545d3f},
  isbn = {978-1-5090-2140-6},
  keywords = {dblp},
  pages = {393-402},
  publisher = {IEEE Computer Society},
  timestamp = {2019-10-17T20:41:53.000+0200},
  title = {High-Performance Hybrid Key-Value Store on Modern Clusters with RDMA Interconnects and SSDs: Non-blocking Extensions, Designs, and Benefits.},
  url = {http://dblp.uni-trier.de/db/conf/ipps/ipdps2016.html#ShankarLIWP16},
  series = {IPDPS '16},
  year = 2016
}

@inproceedings{conf-cloudcom-GugnaniLP16,
  added-at = {2019-02-05T00:00:00.000+0100},
  author = {Gugnani, Shashank and Lu, Xiaoyi and Panda, Dhabaleswar K.},
  biburl = {https://www.bibsonomy.org/bibtex/2a8493ba7d087bf30df51f56edf782146/dblp},
  booktitle = {Proceedings of IEEE International Conference on Cloud Computing Technology and Science},
  crossref = {conf/cloudcom/2016},
  ee = {http://doi.ieeecomputersociety.org/10.1109/CloudCom.2016.0037},
  interhash = {6576e6150f1e1622ea7a7f412d5ddd8c},
  intrahash = {a8493ba7d087bf30df51f56edf782146},
  isbn = {978-1-5090-1445-3},
  keywords = {dblp},
  pages = {152-159},
  publisher = {IEEE Computer Society},
  timestamp = {2019-10-17T19:12:27.000+0200},
  title = {Designing Virtualization-Aware and Automatic Topology Detection Schemes for Accelerating Hadoop on SR-IOV-Enabled Clouds.},
  url = {http://dblp.uni-trier.de/db/conf/cloudcom/cloudcom2016.html#GugnaniLP16},
  series = {CloudCom '16},
  year = 2016
}

@inproceedings{conf-europar-ZhangLCP16,
  added-at = {2018-04-22T00:00:00.000+0200},
  author = {Zhang, Jie and Lu, Xiaoyi and Chakraborty, Sourav and Panda, Dhabaleswar K.},
  biburl = {https://www.bibsonomy.org/bibtex/23f8f46875ce3e5c621b1c81bf17b2811/dblp},
  booktitle = {Proceedings of International European Conference on Parallel Processing},
  crossref = {conf/europar/2016},
  editor = {Dutot, Pierre-François and Trystram, Denis},
  ee = {https://doi.org/10.1007/978-3-319-43659-3_26},
  interhash = {2286f0e065b2760dc8a08d7ffc306ecc},
  intrahash = {3f8f46875ce3e5c621b1c81bf17b2811},
  isbn = {978-3-319-43658-6},
  keywords = {dblp},
  pages = {349-362},
  publisher = {Springer},
  timestamp = {2019-05-15T14:02:55.000+0200},
  title = {Slurm-V: Extending Slurm for Building Efficient HPC Cloud with SR-IOV and IVShmem.},
  url = {http://dblp.uni-trier.de/db/conf/europar/europar2016.html#ZhangLCP16},
  volume = 9833,
  series = {Euro-Par '16},
  year = 2016
}

@inproceedings{conf-icpp-ZhangLP16,
  added-at = {2018-04-22T00:00:00.000+0200},
  author = {Zhang, Jie and Lu, Xiaoyi and Panda, Dhabaleswar K.},
  biburl = {https://www.bibsonomy.org/bibtex/26e7648a41b2defe0664194ca9a5ba8dc/dblp},
  booktitle = {Proceedings of International Conference on Parallel Processing},
  crossref = {conf/icpp/2016},
  ee = {http://doi.ieeecomputersociety.org/10.1109/ICPP.2016.38},
  interhash = {b439fa518e16923c02e0314d061be72b},
  intrahash = {6e7648a41b2defe0664194ca9a5ba8dc},
  isbn = {978-1-5090-2823-8},
  keywords = {dblp},
  pages = {268-277},
  publisher = {IEEE Computer Society},
  timestamp = {2018-04-24T11:39:49.000+0200},
  title = {High Performance MPI Library for Container-Based HPC Cloud on InfiniBand Clusters.},
  url = {http://dblp.uni-trier.de/db/conf/icpp/icpp2016.html#ZhangLP16},
  series = {ICPP '16},
  year = 2016
}

@inproceedings{conf-sc-LiHLSZP16,
  added-at = {2018-04-22T00:00:00.000+0200},
  author = {Li, Mingzhe and Hamidouche, Khaled and Lu, Xiaoyi and Subramoni, Hari and Zhang, Jie and Panda, Dhabaleswar K.},
  biburl = {https://www.bibsonomy.org/bibtex/2a5b2b2866befbe384e395d7318ec0176/dblp},
  booktitle = {Proceedings of International Conference for High Performance Computing, Networking, Storage and Analysis},
  crossref = {conf/sc/2016},
  editor = {West, John and Pancake, Cherri M.},
  ee = {http://dl.acm.org/citation.cfm?id=3014954},
  interhash = {d33688fddca39b8537b18ec9ca542ff3},
  intrahash = {a5b2b2866befbe384e395d7318ec0176},
  isbn = {978-1-4673-8815-3},
  keywords = {dblp},
  pages = {433-443},
  publisher = {IEEE Computer Society},
  timestamp = {2018-04-24T11:41:24.000+0200},
  title = {Designing MPI library with on-demand paging (ODP) of infiniband: challenges and benefits.},
  url = {http://dblp.uni-trier.de/db/conf/sc/sc2016.html#LiHLSZP16},
  series = {SC '16},
  year = 2016
}

@inproceedings{conf-ics-IslamWLP16,
  added-at = {2018-11-06T00:00:00.000+0100},
  author = {Islam, Nusrat Sharmin and ur Rahman, Md. Wasi and Lu, Xiaoyi and Panda, Dhabaleswar K.},
  biburl = {https://www.bibsonomy.org/bibtex/25acb28df0441830a06c42ec51687b3e3/dblp},
  booktitle = {Proceedings of International Conference on Supercompuing},
  crossref = {conf/ics/2016},
  editor = {Ozturk, Ozcan and Ebcioglu, Kemal and Kandemir, Mahmut T. and Mutlu, Onur},
  ee = {https://doi.org/10.1145/2925426.2926290},
  interhash = {e9d30374b6151900d579ccd3985ea5fc},
  intrahash = {5acb28df0441830a06c42ec51687b3e3},
  isbn = {978-1-4503-4361-9},
  keywords = {dblp},
  pages = {8:1-8:14},
  publisher = {ACM},
  timestamp = {2018-11-07T14:15:53.000+0100},
  title = {High Performance Design for HDFS with Byte-Addressability of NVM and RDMA.},
  url = {http://dblp.uni-trier.de/db/conf/ics/ics2016.html#IslamWLP16},
  series = {ICS '16},
  year = 2016
}

@article{journals-tjs-ShankarLWIP16,
	Abstract = {With the emergence of high-performance data analytics, the Hadoop platform is being increasingly used to process data stored on high-performance computing clusters. While there is immense scope for improving the performance of Hadoop MapReduce (including the network-intensive shuffle phase) over these modern clusters, that are equipped with high-speed interconnects such as InfiniBand and 10/40 GigE, and storage systems such as SSDs and Lustre, it is essential to study the MapReduce component in an isolated manner. In this paper, we study popular MapReduce workloads, obtained from well-accepted, comprehensive benchmark suites, to identify common shuffle data distribution patterns. We determine different environmental and workload-specific factors that affect the performance of the MapReduce job. Based on these characterization studies, we propose a micro-benchmark suite that can be used to evaluate the performance of stand-alone Hadoop MapReduce, and demonstrate its ease-of-use with different networks/protocols, Hadoop distributions, and storage architectures. Performance evaluations with our proposed micro-benchmarks show that stand-alone Hadoop MapReduce over IPoIB performs better than 10 GigE by about 13--15 {\%}, and the RDMA-enhanced hybrid MapReduce design can achieve up to 43 {\%} performance improvement over default Hadoop MapReduce over IPoIB, in both shared-nothing and shared storage architectures.},
	Author = {Shankar, Dipti and Lu, Xiaoyi and Wasi-ur-Rahman, Md. and Islam, Nusrat and Panda, Dhabaleswar K.},
	Da = {2016/12/01},
	Date-Added = {2021-02-03 03:52:03 +0000},
	Date-Modified = {2021-02-03 03:52:03 +0000},
	Doi = {10.1007/s11227-016-1760-5},
	Id = {Shankar2016},
	Isbn = {1573-0484},
	Journal = {The Journal of Supercomputing},
	Number = {12},
	Pages = {4573--4600},
	Title = {Characterizing and benchmarking stand-alone Hadoop MapReduce on modern HPC clusters},
	Ty = {JOUR},
	Url = {https://doi.org/10.1007/s11227-016-1760-5},
	Volume = {72},
	Year = {2016},
	series = {JSC '16},
	Bdsk-Url-1 = {https://doi.org/10.1007/s11227-016-1760-5}}

@inproceedings{conf-bpoe-kunal,
  author={Kulkarni, Kunal and  Lu, Xiayi and Panda, Dhabaleswar K.},
  title={Characterizing Cloudera Impala Workloads with BigDataBench on InfiniBand Clusters},
  booktitle={Proceedings of The 7th International Workshop on Big Data Benchmarks, Performance, Optimization, and Emerging Hardware (BPOE-7)},
  year={2016},
  month={April},
  location={ Atlanta, Georgia },
  series = {BPOE '16}
}

@inproceedings{conf-bigdataconf-ShankarLP16,
  added-at = {2017-05-26T00:00:00.000+0200},
  author = {Shankar, Dipti and Lu, Xiaoyi and Panda, Dhabaleswar K.},
  biburl = {https://www.bibsonomy.org/bibtex/24c62042a0b12a71a4dc8a9a1418782b2/dblp},
  booktitle = {Proceedings of IEEE International Conference on Big Data},
  crossref = {conf/bigdataconf/2016},
  editor = {Joshi, James and Karypis, George and Liu, Ling and Hu, Xiaohua and Ak, Ronay and Xia, Yinglong and Xu, Weijia and Sato, Aki-Hiro and Rachuri, Sudarsan and Ungar, Lyle H. and Yu, Philip S. and Govindaraju, Rama and Suzumura, Toyotaro},
  ee = {https://doi.org/10.1109/BigData.2016.7840630},
  interhash = {5eb626a4d0dd04f76f5b0eb291dad47b},
  intrahash = {4c62042a0b12a71a4dc8a9a1418782b2},
  isbn = {978-1-4673-9005-7},
  keywords = {dblp},
  pages = {404-409},
  publisher = {IEEE Computer Society},
  timestamp = {2019-02-21T11:45:26.000+0100},
  title = {Boldio: A hybrid and resilient burst-buffer over lustre for accelerating big data I/O.},
  url = {http://dblp.uni-trier.de/db/conf/bigdataconf/bigdataconf2016.html#ShankarLP16},
  series = {BigData '16},
  note = {Short Paper},
  year = 2016
}

@inproceedings{conf-cloudcom-LuSGSP16,
  added-at = {2019-02-05T00:00:00.000+0100},
  author = {Lu, Xiaoyi and Shankar, Dipti and Gugnani, Shashank and Subramoni, Hari and Panda, Dhabaleswar K.},
  biburl = {https://www.bibsonomy.org/bibtex/2a46a339e7117d4a06bc7d32d1a300037/dblp},
  booktitle = {Proceedings of IEEE International Conference on Cloud Computing Technology and Science},
  crossref = {conf/cloudcom/2016},
  ee = {http://doi.ieeecomputersociety.org/10.1109/CloudCom.2016.0057},
  interhash = {7b80a78af7638775e3fb7c0467e66790},
  intrahash = {a46a339e7117d4a06bc7d32d1a300037},
  isbn = {978-1-5090-1445-3},
  keywords = {dblp},
  pages = {310-317},
  publisher = {IEEE Computer Society},
  timestamp = {2019-10-17T19:12:27.000+0200},
  title = {Impact of HPC Cloud Networking Technologies on Accelerating Hadoop RPC and HBase.},
  url = {http://dblp.uni-trier.de/db/conf/cloudcom/cloudcom2016.html#LuSGSP16},
  series = {CloudCom '16},
  year = 2016
}

@inproceedings{conf-xsede-TatineniLCMP16,
  added-at = {2018-11-09T00:00:00.000+0100},
  author = {Tatineni, Mahidhar and Lu, Xiaoyi and Choi, Dong Ju and Majumdar, Amitava and Panda, Dhabaleswar K.},
  biburl = {https://www.bibsonomy.org/bibtex/2ba3ffd7c708a834d4e9f554127242dca/dblp},
  booktitle = {Proceedings of Annual Conference on Extreme Science and Engineering Discovery Environment},
  crossref = {conf/xsede/2016},
  ee = {https://doi.org/10.1145/2949550.2949561},
  interhash = {06716e6dc8abcda51a12d3a559d4fe6a},
  intrahash = {ba3ffd7c708a834d4e9f554127242dca},
  isbn = {978-1-4503-4755-6},
  keywords = {dblp},
  pages = {23:1-23:5},
  publisher = {ACM},
  timestamp = {2018-11-10T11:41:33.000+0100},
  title = {Experiences and Benefits of Running RDMA Hadoop and Spark on SDSC Comet.},
  url = {http://dblp.uni-trier.de/db/conf/xsede/xsede2016.html#TatineniLCMP16},
  series = {XSEDE '16},
  year = 2016
}

@inproceedings{conf-bigdataconf-IslamWLP16,
  added-at = {2019-02-20T00:00:00.000+0100},
  author = {Islam, Nusrat Sharmin and ur Rahman, Md. Wasi and Lu, Xiaoyi and Panda, Dhabaleswar K.},
  biburl = {https://www.bibsonomy.org/bibtex/232e4d44e53f16991fc53c4936865b796/dblp},
  booktitle = {Proceedings of IEEE International Conference on Big Data},
  crossref = {conf/bigdataconf/2016},
  editor = {Joshi, James and Karypis, George and Liu, Ling and Hu, Xiaohua and Ak, Ronay and Xia, Yinglong and Xu, Weijia and Sato, Aki-Hiro and Rachuri, Sudarsan and Ungar, Lyle H. and Yu, Philip S. and Govindaraju, Rama and Suzumura, Toyotaro},
  ee = {http://doi.ieeecomputersociety.org/10.1109/BigData.2016.7840608},
  interhash = {69169d2461aa916baa911ec02b8cc553},
  intrahash = {32e4d44e53f16991fc53c4936865b796},
  isbn = {978-1-4673-9005-7},
  keywords = {dblp},
  pages = {223-232},
  publisher = {IEEE Computer Society},
  timestamp = {2019-02-21T11:45:26.000+0100},
  title = {Efficient data access strategies for Hadoop and Spark on HPC cluster with heterogeneous storage.},
  url = {http://dblp.uni-trier.de/db/conf/bigdataconf/bigdataconf2016.html#IslamWLP16},
  series = {BigData '16},
  year = 2016
}

@incollection{books-sp-16-Lu0P16,
author="Lu, Xiaoyi
and Wasi-ur-Rahman, Md.
and Islam, Nusrat
and Shankar, Dipti
and Panda, Dhabaleswar K. (DK)",
editor="Arora, Ritu",
title="Accelerating Big Data Processing on Modern HPC Clusters",
bookTitle="Conquering Big Data with High Performance Computing",
year="2016",
publisher="Springer International Publishing",
address="Cham",
pages="81--107",
abstract="Modern HPC systems and the associated middleware (such as MPI and parallel file systems) have been exploiting the advances in HPC technologies (multi-/many-core architecture, RDMA-enabled networking, and SSD) for many years. However, Big Data processing and management middleware have not fully taken advantage of such technologies. These disparities are taking HPC and Big Data processing into divergent trajectories. This chapter provides an overview of popular Big Data processing middleware, high-performance interconnects and storage architectures, and discusses the challenges in accelerating Big Data processing middleware by leveraging emerging technologies on modern HPC clusters. This chapter presents case studies of advanced designs based on RDMA and heterogeneous storage architecture, that were proposed to address these challenges for multiple components of Hadoop (HDFS and MapReduce) and Spark. The advanced designs presented in the case studies are publicly available as a part of the High-Performance Big Data (HiBD) project. An overview of the HiBD project is also provided in this chapter. All these works aim to bring HPC and Big Data processing into a convergent trajectory.",
isbn="978-3-319-33742-5",
doi="10.1007/978-3-319-33742-5_5",
series="Chapter '16",
url="https://doi.org/10.1007/978-3-319-33742-5_5"
}

@inproceedings{conf-ipps-ZhangLP16,
  added-at = {2018-04-22T00:00:00.000+0200},
  author = {Zhang, Jie and Lu, Xiaoyi and Panda, Dhabaleswar K.},
  biburl = {https://www.bibsonomy.org/bibtex/2d542680a0dffbbe092f4a954d1cdd8c1/dblp},
  booktitle = {Workshop Proceedings of IEEE International Parallel and Distributed Processing Symposium},
  crossref = {conf/ipps/2016w},
  ee = {http://doi.ieeecomputersociety.org/10.1109/IPDPSW.2016.178},
  interhash = {bca70fc9fbc30ea427fb3ffb004a6045},
  intrahash = {d542680a0dffbbe092f4a954d1cdd8c1},
  isbn = {978-1-5090-3682-0},
  keywords = {dblp},
  pages = {1777-1784},
  publisher = {IEEE Computer Society},
  timestamp = {2019-10-17T20:33:40.000+0200},
  title = {Performance Characterization of Hypervisor-and Container-Based Virtualization for HPC on SR-IOV Enabled InfiniBand Clusters.},
  url = {http://dblp.uni-trier.de/db/conf/ipps/ipdps2016w.html#ZhangLP16},
  series = {IPDPS Workshops '16},
  year = 2016
}

@inproceedings{conf-supercomputer-SubramoniAAPLHP16,
  added-at = {2017-05-23T00:00:00.000+0200},
  author = {Subramoni, Hari and Augustine, Albert Mathews and Arnold, Mark Daniel and Perkins, Jonathan L. and Lu, Xiaoyi and Hamidouche, Khaled and Panda, Dhabaleswar K.},
  biburl = {https://www.bibsonomy.org/bibtex/2ea975cde67cd404aface8e93a0a77ea5/dblp},
  booktitle = {Proceedings of the International Supercomputing Conference},
  crossref = {conf/supercomputer/2016},
  editor = {Kunkel, Julian M. and Balaji, Pavan and Dongarra, Jack J.},
  ee = {https://doi.org/10.1007/978-3-319-41321-1_16},
  interhash = {6d32b3752f860781e8338ae7e88cf1e3},
  intrahash = {ea975cde67cd404aface8e93a0a77ea5},
  isbn = {978-3-319-41320-4},
  keywords = {dblp},
  pages = {300-320},
  publisher = {Springer},
  timestamp = {2017-08-22T11:55:17.000+0200},
  title = {INAM2: InfiniBand Network Analysis and Monitoring with MPI.},
  url = {http://dblp.uni-trier.de/db/conf/supercomputer/isc2016.html#SubramoniAAPLHP16},
  volume = 9697,
  series = {ISC '16},
  year = 2016
}

@inproceedings{conf-hipc-LiLHZP16,
  added-at = {2019-02-07T00:00:00.000+0100},
  author = {Li, Mingzhe and Lu, Xiaoyi and Hamidouche, Khaled and Zhang, Jie and Panda, Dhabaleswar K.},
  biburl = {https://www.bibsonomy.org/bibtex/28a18dad70b817fd5a01733841ad80d84/dblp},
  booktitle = {Proceedings of IEEE International Conference on High Performance Computing},
  crossref = {conf/hipc/2016},
  ee = {http://doi.ieeecomputersociety.org/10.1109/HiPC.2016.015},
  interhash = {518744da45a2e013c81c79014cca7dad},
  intrahash = {8a18dad70b817fd5a01733841ad80d84},
  isbn = {978-1-5090-5411-4},
  keywords = {dblp},
  pages = {42-51},
  publisher = {IEEE Computer Society},
  timestamp = {2019-10-17T16:01:19.000+0200},
  title = {Mizan-RMA: Accelerating Mizan Graph Processing Framework with MPI RMA.},
  url = {http://dblp.uni-trier.de/db/conf/hipc/hipc2016.html#LiLHZP16},
  series = {HiPC '16},
  year = 2016
}

@inproceedings{conf-sc-Wasi-ur-RahmanI16,
  added-at = {2019-02-14T00:00:00.000+0100},
  author = {ur Rahman, Md. Wasi and Islam, Nusrat Sharmin and Lu, Xiaoyi and Panda, Dhabaleswar K.},
  biburl = {https://www.bibsonomy.org/bibtex/2df40c1588004a2d84971acb2acfee616/dblp},
  booktitle = {Proceedings of Joint International Workshop on Parallel Data Storage and Data Intensive Scalable Computing Systems},
  crossref = {conf/sc/2016pdswdiscs},
  ee = {http://dl.acm.org/citation.cfm?id=3019050},
  interhash = {385f486f1eeb07fee38b316561e9a370},
  intrahash = {df40c1588004a2d84971acb2acfee616},
  isbn = {978-1-5090-5216-5},
  keywords = {dblp},
  pages = {19-24},
  publisher = {IEEE Computer Society},
  timestamp = {2019-02-15T11:42:25.000+0100},
  title = {Can Non-volatile Memory Benefit MapReduce Applications on HPC Clusters?},
  url = {http://dblp.uni-trier.de/db/conf/sc/pdsw-discs2016.html#Wasi-ur-RahmanI16},
  series = {PDSW-DISCS '16},
  year = 2016
}

@inproceedings{conf-ccgrid-ZhangLAP15,
  added-at = {2018-04-22T00:00:00.000+0200},
  author = {Zhang, Jie and Lu, Xiaoyi and Arnold, Mark Daniel and Panda, Dhabaleswar K.},
  biburl = {https://www.bibsonomy.org/bibtex/2d7daf9a4e7b650e1b034249ffdf7ee2d/dblp},
  booktitle = {Proceedings of IEEE/ACM International Symposium on Cluster, Cloud and Grid Computing},
  crossref = {conf/ccgrid/2015},
  ee = {http://doi.ieeecomputersociety.org/10.1109/CCGrid.2015.166},
  interhash = {30aeee0d434d4bfabf9067001d0903d2},
  intrahash = {d7daf9a4e7b650e1b034249ffdf7ee2d},
  isbn = {978-1-4799-8006-2},
  keywords = {dblp},
  pages = {71-80},
  publisher = {IEEE Computer Society},
  timestamp = {2018-04-24T11:41:52.000+0200},
  title = {MVAPICH2 over OpenStack with SR-IOV: An Efficient Approach to Build HPC Clouds.},
  url = {http://dblp.uni-trier.de/db/conf/ccgrid/ccgrid2015.html#ZhangLAP15},
  series = {CCGRID '15},
  year = 2015
}

@inproceedings{conf-cluster-LiSHLP15,
  added-at = {2016-04-27T00:00:00.000+0200},
  author = {Li, Mingzhe and Subramoni, Hari and Hamidouche, Khaled and Lu, Xiaoyi and Panda, Dhabaleswar K.},
  biburl = {https://www.bibsonomy.org/bibtex/2fd06f6c2a0f1f5077dca137168456144/dblp},
  booktitle = {Proceedings of IEEE International Conference on Cluster Computing},
  crossref = {conf/cluster/2015},
  ee = {http://doi.ieeecomputersociety.org/10.1109/CLUSTER.2015.41},
  interhash = {f7a9d26ad423b0acf1d90d12643b9859},
  intrahash = {fd06f6c2a0f1f5077dca137168456144},
  isbn = {978-1-4673-6598-7},
  keywords = {dblp},
  pages = {226-235},
  publisher = {IEEE Computer Society},
  timestamp = {2016-04-28T11:37:02.000+0200},
  title = {High Performance MPI Datatype Support with User-Mode Memory Registration: Challenges, Designs, and Benefits.},
  url = {http://dblp.uni-trier.de/db/conf/cluster/cluster2015.html#LiSHLP15},
  series = {CLUSTER '15},
  year = 2015
}

@inproceedings{conf-ccgrid-IslamLWSP15,
  added-at = {2016-01-20T00:00:00.000+0100},
  author = {Islam, Nusrat Sharmin and Lu, Xiaoyi and ur Rahman, Md. Wasi and Shankar, Dipti and Panda, Dhabaleswar K.},
  biburl = {https://www.bibsonomy.org/bibtex/25f33010fc10f5d31f8308be2f3efa8ce/dblp},
  booktitle = {Proceedings of IEEE/ACM International Symposium on Cluster, Cloud and Grid Computing},
  crossref = {conf/ccgrid/2015},
  ee = {http://doi.ieeecomputersociety.org/10.1109/CCGrid.2015.161},
  interhash = {d4086d233affa7e2b9935d513a0476b7},
  intrahash = {5f33010fc10f5d31f8308be2f3efa8ce},
  isbn = {978-1-4799-8006-2},
  keywords = {dblp},
  pages = {101-110},
  publisher = {IEEE Computer Society},
  timestamp = {2016-01-21T11:48:38.000+0100},
  title = {Triple-H: A Hybrid Approach to Accelerate HDFS on HPC Clusters with Heterogeneous Storage Architecture.},
  url = {http://dblp.uni-trier.de/db/conf/ccgrid/ccgrid2015.html#IslamLWSP15},
  series = {CCGRID '15},
  year = 2015
}

@inproceedings{conf-bigdataconf-ShankarLWIP15,
  added-at = {2018-11-03T00:00:00.000+0100},
  author = {Shankar, Dipti and Lu, Xiaoyi and ur Rahman, Md. Wasi and Islam, Nusrat S. and Panda, Dhabaleswar K.},
  biburl = {https://www.bibsonomy.org/bibtex/22457ab17750ef4518b7803012d725d4b/dblp},
  booktitle = {Proceedings of IEEE International Conference on Big Data},
  crossref = {conf/bigdataconf/2015},
  ee = {https://doi.org/10.1109/BigData.2015.7363797},
  interhash = {58673299d42c4d0421f7e462b759e321},
  intrahash = {2457ab17750ef4518b7803012d725d4b},
  isbn = {978-1-4799-9926-2},
  keywords = {dblp},
  pages = {539-544},
  publisher = {IEEE Computer Society},
  timestamp = {2019-02-21T11:47:24.000+0100},
  title = {Benchmarking key-value stores on high-performance storage and interconnects for web-scale workloads.},
  url = {http://dblp.uni-trier.de/db/conf/bigdataconf/bigdataconf2015.html#ShankarLWIP15},
  series = {BigData '15},
  note = {Short Paper},
  year = 2015
}

@inproceedings{conf-icdcs-ChaoLLLX15,
  added-at = {2020-04-08T00:00:00.000+0200},
  author = {Chao, Lu and Li, Chundian and Liang, Fan and Lu, Xiaoyi and Xu, Zhiwei},
  biburl = {https://www.bibsonomy.org/bibtex/28a30fc5c004d37dffe977167f500efdc/dblp},
  booktitle = {Proceedings of IEEE International Conference on Distributed Computing Systems},
  crossref = {conf/icdcs/2015},
  ee = {http://doi.ieeecomputersociety.org/10.1109/ICDCS.2015.73},
  interhash = {d2452629ecf13b87cdf932e50d4ef4cf},
  intrahash = {8a30fc5c004d37dffe977167f500efdc},
  isbn = {978-1-4673-7214-5},
  keywords = {dblp},
  pages = {664-673},
  publisher = {IEEE Computer Society},
  timestamp = {2020-04-09T11:42:03.000+0200},
  title = {Accelerating Apache Hive with MPI for Data Warehouse Systems.},
  url = {http://dblp.uni-trier.de/db/conf/icdcs/icdcs2015.html#ChaoLLLX15},
  series = {ICDCS '15},
  year = 2015
}

@inproceedings{conf-bigdataservice-LinLLZX15,
  added-at = {2020-09-05T00:00:00.000+0200},
  author = {Lin, Jian and Liang, Fan and Lu, Xiaoyi and Zha, Li and Xu, Zhiwei},
  biburl = {https://www.bibsonomy.org/bibtex/29cac0e3b019fc3d478bb0bed948b973d/dblp},
  booktitle = {Proceedings of IEEE International Conference on Big Data Computing Service and Applications},
  crossref = {conf/bigdataservice/2015},
  ee = {http://doi.ieeecomputersociety.org/10.1109/BigDataService.2015.33},
  interhash = {5eff703b1bdbcf6776c819db6716f993},
  intrahash = {9cac0e3b019fc3d478bb0bed948b973d},
  isbn = {978-1-4799-8128-1},
  keywords = {dblp},
  pages = {176-183},
  publisher = {IEEE Computer Society},
  timestamp = {2020-09-09T14:40:51.000+0200},
  title = {Modeling and Designing Fault-Tolerance Mechanisms for MPI-Based MapReduce Data Computing Framework.},
  url = {http://dblp.uni-trier.de/db/conf/bigdataservice/bigdataservice2015.html#LinLLZX15},
  series = {BigDataService '15},
  note = {Short Paper},
  year = 2015
}

@article{journals-jcst-LiangL15,
	Abstract = {Current popular systems, Hadoop and Spark, cannot achieve satisfied performance because of the inefficient overlapping of computation and communication when running iterative big data applications. The pipeline of computing, data movement, and data management plays a key role for current distributed data computing systems. In this paper, we first analyze the overhead of shuffle operation in Hadoop and Spark when running PageRank workload, and then propose an event-driven pipeline and in-memory shuffle design with better overlapping of computation and communication as DataMPI-Iteration, an MPI-based library, for iterative big data computing. Our performance evaluation shows DataMPI-Iteration can achieve 9X∼21X speedup over Apache Hadoop, and 2X∼3X speedup over Apache Spark for PageRank and K-means.},
	Author = {Liang, Fan and Lu, Xiaoyi},
	Da = {2015/03/01},
	Date-Added = {2021-02-03 03:39:54 +0000},
	Date-Modified = {2021-02-03 03:39:54 +0000},
	Doi = {10.1007/s11390-015-1522-5},
	Id = {Liang2015},
	Isbn = {1860-4749},
	Journal = {Journal of Computer Science and Technology},
	Number = {2},
	Pages = {283--294},
	Title = {Accelerating Iterative Big Data Computing Through MPI},
	Ty = {JOUR},
	Url = {https://doi.org/10.1007/s11390-015-1522-5},
	Volume = {30},
	Year = {2015},
	series = {JCST '15},
	Bdsk-Url-1 = {https://doi.org/10.1007/s11390-015-1522-5}}

@inproceedings{conf-ipps-LinHLLP15,
  added-at = {2020-09-05T00:00:00.000+0200},
  author = {Lin, Jian and Hamidouche, Khaled and Lu, Xiaoyi and Li, Mingzhe and Panda, Dhabaleswar K.},
  biburl = {https://www.bibsonomy.org/bibtex/2f5c3a59a348cfd36b98d62e7bc9233e2/dblp},
  booktitle = {Workshop Proceedings of IEEE International Parallel and Distributed Processing Symposium},
  crossref = {conf/ipps/2015w},
  ee = {http://doi.ieeecomputersociety.org/10.1109/IPDPSW.2015.115},
  interhash = {35a0ae2630d56e70726f7a6f1df50fc6},
  intrahash = {f5c3a59a348cfd36b98d62e7bc9233e2},
  isbn = {978-1-4673-7684-6},
  keywords = {dblp},
  pages = {225-234},
  publisher = {IEEE Computer Society},
  timestamp = {2020-09-09T13:57:41.000+0200},
  title = {High-Performance Coarray Fortran Support with MVAPICH2-X: Initial Experience and Evaluation.},
  url = {http://dblp.uni-trier.de/db/conf/ipps/ipdps2015w.html#LinHLLP15},
  series = {IPDPS Workshops '15},
  year = 2015
}

@inproceedings{conf-openshmem-LinHZLVP15,
  added-at = {2020-09-05T00:00:00.000+0200},
  author = {Lin, Jian and Hamidouche, Khaled and Zhang, Jie and Lu, Xiaoyi and Vishnu, Abhinav and Panda, Dhabaleswar K.},
  biburl = {https://www.bibsonomy.org/bibtex/2c7d94be818b10fb98f54e2568792f615/dblp},
  booktitle = {Proceedings of Workshop on OpenSHMEM and Related Technologies},
  crossref = {conf/openshmem/2015},
  editor = {Venkata, Manjunath Gorentla and Shamis, Pavel and Imam, Neena and Lopez, M. Graham},
  ee = {https://doi.org/10.1007/978-3-319-26428-8_11},
  interhash = {33262b17b87b6540f8dd3b1c9507f597},
  intrahash = {c7d94be818b10fb98f54e2568792f615},
  isbn = {978-3-319-26427-1},
  keywords = {dblp},
  pages = {164-177},
  publisher = {Springer},
  timestamp = {2020-09-09T14:13:41.000+0200},
  title = {Accelerating k-NN Algorithm with Hybrid MPI and OpenSHMEM.},
  url = {http://dblp.uni-trier.de/db/conf/openshmem/openshmem2015.html#LinHZLVP15},
  volume = 9397,
  series = {OpenSHMEM '15},
  year = 2015
}

@inproceedings{conf-europar-LiHLLP15,
  added-at = {2017-05-24T00:00:00.000+0200},
  author = {Li, Mingzhe and Hamidouche, Khaled and Lu, Xiaoyi and Lin, Jian and Panda, Dhabaleswar K.},
  biburl = {https://www.bibsonomy.org/bibtex/2a9f141c5e856df7a6a4199c7ee6349a5/dblp},
  booktitle = {Proceedings of International European Conference on Parallel Processing},
  crossref = {conf/europar/2015},
  editor = {Träff, Jesper Larsson and Hunold, Sascha and Versaci, Francesco},
  ee = {https://doi.org/10.1007/978-3-662-48096-0_48},
  interhash = {3feb369f715d8acfbbfb04ce1b2a4bf7},
  intrahash = {a9f141c5e856df7a6a4199c7ee6349a5},
  isbn = {978-3-662-48095-3},
  keywords = {dblp},
  pages = {625-637},
  publisher = {Springer},
  timestamp = {2019-05-15T14:04:55.000+0200},
  title = {High-Performance and Scalable Design of MPI-3 RMA on Xeon Phi Clusters.},
  url = {http://dblp.uni-trier.de/db/conf/europar/europar2015.html#LiHLLP15},
  volume = 9233,
  series = {Euro-Par '15},
  year = 2015
}

@inproceedings{conf-hipc-LiHLZLP15,
  added-at = {2018-04-22T00:00:00.000+0200},
  author = {Li, Mingzhe and Hamidouche, Khaled and Lu, Xiaoyi and Zhang, Jie and Lin, Jian and Panda, Dhabaleswar K.},
  biburl = {https://www.bibsonomy.org/bibtex/25b850a40647848a193d06759206e840b/dblp},
  booktitle = {Proceedings of IEEE International Conference on High Performance Computing},
  crossref = {conf/hipc/2015},
  ee = {http://doi.ieeecomputersociety.org/10.1109/HiPC.2015.48},
  interhash = {5600be30134880e2629e34320a99429f},
  intrahash = {5b850a40647848a193d06759206e840b},
  isbn = {978-1-4673-8488-9},
  keywords = {dblp},
  pages = {244-253},
  publisher = {IEEE Computer Society},
  timestamp = {2019-10-17T16:01:12.000+0200},
  title = {High Performance OpenSHMEM Strided Communication Support with InfiniBand UMR.},
  url = {http://dblp.uni-trier.de/db/conf/hipc/hipc2015.html#LiHLZLP15},
  series = {HiPC '15},
  year = 2015
}

@inproceedings{conf-bpoe-BhatILWSP15,
  added-at = {2017-05-24T00:00:00.000+0200},
  author = {Bhat, Adithya and Islam, Nusrat Sharmin and Lu, Xiaoyi and ur Rahman, Md. Wasi and Shankar, Dipti and Panda, Dhabaleswar K.},
  biburl = {https://www.bibsonomy.org/bibtex/2b7df9c408d3a045fc2a3996650225bc0/dblp},
  booktitle = {Proceedings of International Workshop on Big Data Benchmarks, Performance Optimization, and Emerging Hardware},
  crossref = {conf/bpoe/2015},
  editor = {Zhan, Jianfeng and Han, Rui and Zicari, Roberto V.},
  ee = {https://doi.org/10.1007/978-3-319-29006-5_10},
  interhash = {c0cec3a21ae6187d5258d66b8954e6a9},
  intrahash = {b7df9c408d3a045fc2a3996650225bc0},
  isbn = {978-3-319-29005-8},
  keywords = {dblp},
  pages = {119-132},
  publisher = {Springer},
  timestamp = {2019-05-15T14:16:42.000+0200},
  title = {A Plugin-Based Approach to Exploit RDMA Benefits for Apache and Enterprise HDFS.},
  url = {http://dblp.uni-trier.de/db/conf/bpoe/bpoe2015.html#BhatILWSP15},
  volume = 9495,
  series = {BPOE '15},
  year = 2015
}

@inproceedings{conf-ipps-Wasi-ur-RahmanL15,
  added-at = {2017-05-24T00:00:00.000+0200},
  author = {ur Rahman, Md. Wasi and Lu, Xiaoyi and Islam, Nusrat Sharmin and Rajachandrasekar, Raghunath and Panda, Dhabaleswar K.},
  biburl = {https://www.bibsonomy.org/bibtex/2e3f3a1a141829ff3fa319a0b27553182/dblp},
  booktitle = {Proceedings of IEEE International Parallel and Distributed Processing Symposium},
  crossref = {conf/ipps/2015},
  ee = {http://doi.ieeecomputersociety.org/10.1109/IPDPS.2015.83},
  interhash = {bec243d576b1f07fcaadccb7810e9bb0},
  intrahash = {e3f3a1a141829ff3fa319a0b27553182},
  isbn = {978-1-4799-8649-1},
  keywords = {dblp},
  pages = {291-300},
  publisher = {IEEE Computer Society},
  timestamp = {2019-10-17T20:39:06.000+0200},
  title = {High-Performance Design of YARN MapReduce on Modern HPC Clusters with Lustre and RDMA.},
  url = {http://dblp.uni-trier.de/db/conf/ipps/ipdps2015.html#Wasi-ur-RahmanL15},
  series = {IPDPS '15},
  year = 2015
}

@inproceedings{conf-icpp-IslamSLWP15,
  added-at = {2015-12-18T00:00:00.000+0100},
  author = {Islam, Nusrat Sharmin and Shankar, Dipti and Lu, Xiaoyi and ur Rahman, Md. Wasi and Panda, Dhabaleswar K.},
  biburl = {https://www.bibsonomy.org/bibtex/2ffddeeb84aa237368c20a8f5b2556656/dblp},
  booktitle = {Proceedings of International Conference on Parallel Processing},
  crossref = {conf/icpp/2015},
  ee = {http://doi.ieeecomputersociety.org/10.1109/ICPP.2015.79},
  interhash = {eadde109be8a26bbfdec304ab730ee74},
  intrahash = {ffddeeb84aa237368c20a8f5b2556656},
  isbn = {978-1-4673-7587-0},
  keywords = {dblp},
  pages = {280-289},
  publisher = {IEEE Computer Society},
  timestamp = {2015-12-24T12:16:30.000+0100},
  title = {Accelerating I/O Performance of Big Data Analytics on HPC Clusters through RDMA-Based Key-Value Store.},
  url = {http://dblp.uni-trier.de/db/conf/icpp/icpp2015.html#IslamSLWP15},
  series = {ICPP '15},
  year = 2015
}

@inproceedings{conf-ispass-ShankarLJWIP15,
  added-at = {2017-05-23T00:00:00.000+0200},
  author = {Shankar, Dipti and Lu, Xiaoyi and Jose, Jithin and ur Rahman, Md. Wasi and Islam, Nusrat S. and Panda, Dhabaleswar K.},
  biburl = {https://www.bibsonomy.org/bibtex/2f699ef0065eb39ae18c86116526099a7/dblp},
  booktitle = {Proceedings of IEEE International Symposium on Performance Analysis of Systems and Software},
  crossref = {conf/ispass/2015},
  ee = {http://doi.ieeecomputersociety.org/10.1109/ISPASS.2015.7095796},
  interhash = {1b36460e57b49f95fe1ab5d67db7453a},
  intrahash = {f699ef0065eb39ae18c86116526099a7},
  isbn = {978-1-4799-1957-4},
  keywords = {dblp},
  pages = {159-160},
  publisher = {IEEE Computer Society},
  timestamp = {2019-10-17T19:16:56.000+0200},
  title = {Can RDMA benefit online data processing workloads on memcached and MySQL?},
  url = {http://dblp.uni-trier.de/db/conf/ispass/ispass2015.html#ShankarLJWIP15},
  series = {ISPASS '15},
  note = {Poster Paper},
  year = 2015
}

@inproceedings{conf-bigdataconf-IslamWLSP15,
  added-at = {2018-11-03T00:00:00.000+0100},
  author = {Islam, Nusrat Sharmin and ur Rahman, Md. Wasi and Lu, Xiaoyi and Shankar, Dipti and Panda, Dhabaleswar K.},
  biburl = {https://www.bibsonomy.org/bibtex/2d34bfb4288a488a5a882bbe6d5db7ec9/dblp},
  booktitle = {Proceedings of IEEE International Conference on Big Data},
  crossref = {conf/bigdataconf/2015},
  ee = {https://doi.org/10.1109/BigData.2015.7363761},
  interhash = {92edfaa9f7780bdd075c8a55c6516abf},
  intrahash = {d34bfb4288a488a5a882bbe6d5db7ec9},
  isbn = {978-1-4799-9926-2},
  keywords = {dblp},
  pages = {243-252},
  publisher = {IEEE Computer Society},
  timestamp = {2019-02-21T11:47:24.000+0100},
  title = {Performance characterization and acceleration of in-memory file systems for Hadoop and Spark applications on HPC clusters.},
  url = {http://dblp.uni-trier.de/db/conf/bigdataconf/bigdataconf2015.html#IslamWLSP15},
  series = {BigData '15},
  year = 2015
}

@inproceedings{conf-bigdataconf-IslamLWRP14,
  added-at = {2017-05-26T00:00:00.000+0200},
  author = {Islam, Nusrat Sharmin and Lu, Xiaoyi and ur Rahman, Md. Wasi and Rajachandrasekar, Raghunath and Panda, Dhabaleswar K.},
  biburl = {https://www.bibsonomy.org/bibtex/20c9489c2fb3ff8be409cb7a823fd7af3/dblp},
  booktitle = {Proceedings of IEEE International Conference on Big Data},
  crossref = {conf/bigdataconf/2014},
  editor = {Lin, Jimmy J. and Pei, Jian and Hu, Xiaohua and Chang, Wo and Nambiar, Raghunath and Aggarwal, Charu C. and Cercone, Nick and Honavar, Vasant G. and Huan, Jun and Mobasher, Bamshad and Pyne, Saumyadipta},
  ee = {https://doi.org/10.1109/BigData.2014.7004235},
  interhash = {f7d1f192ac4866428d9b9a8739439ce2},
  intrahash = {0c9489c2fb3ff8be409cb7a823fd7af3},
  isbn = {978-1-4799-5665-4},
  keywords = {dblp},
  pages = {213-218},
  publisher = {IEEE},
  timestamp = {2017-12-07T22:56:35.000+0100},
  title = {In-memory I/O and replication for HDFS with Memcached: Early experiences.},
  url = {http://dblp.uni-trier.de/db/conf/bigdataconf/bigdataconf2014.html#IslamLWRP14},
  series = {BigData '14},
  note = {Short Paper},
  year = 2014
}

@inproceedings{conf-pgas-LiLLHTP14,
  added-at = {2018-11-06T00:00:00.000+0100},
  author = {Li, Mingzhe and Lin, Jian and Lu, Xiaoyi and Hamidouche, Khaled and Tomko, Karen and Panda, Dhabaleswar K.},
  biburl = {https://www.bibsonomy.org/bibtex/2e592759c72899858f4e648c7c96f1d31/dblp},
  booktitle = {Proceedings of International Conference on Partitioned Global Address Space Programming Models},
  crossref = {conf/pgas/2014},
  editor = {Malony, Allen D. and Hammond, Jeff R.},
  ee = {https://doi.org/10.1145/2676870.2676893},
  interhash = {a8019f7ad2a1aee25237457fb32a1e81},
  intrahash = {e592759c72899858f4e648c7c96f1d31},
  isbn = {978-1-4503-3247-7},
  keywords = {dblp},
  pages = {24:1-24:4},
  publisher = {ACM},
  timestamp = {2018-11-07T14:54:20.000+0100},
  title = {Scalable MiniMD Design with Hybrid MPI and OpenSHMEM.},
  url = {http://dblp.uni-trier.de/db/conf/pgas/pgas2014.html#LiLLHTP14},
  series = {PGAS '14},
  year = 2014
}

@inproceedings{conf-ppopp-LuoLHKP14,
  added-at = {2014-02-09T00:00:00.000+0100},
  author = {Luo, Miao and Lu, Xiaoyi and Hamidouche, Khaled and Kandalla, Krishna Chaitanya and Panda, Dhabaleswar K.},
  biburl = {https://www.bibsonomy.org/bibtex/2e5871614e3398d9a770dc1f541070874/dblp},
  booktitle = {Proceedings of ACM SIGPLAN Symposium on Principles and Practice of Parallel Programming},
  crossref = {conf/ppopp/2014},
  editor = {Moreira, José E. and Larus, James R.},
  ee = {http://doi.acm.org/10.1145/2555243.2555287},
  interhash = {eedac88816d558415ddb23954f8c9207},
  intrahash = {e5871614e3398d9a770dc1f541070874},
  isbn = {978-1-4503-2656-8},
  keywords = {dblp},
  pages = {395-396},
  publisher = {ACM},
  timestamp = {2016-07-05T11:40:06.000+0200},
  title = {Initial study of multi-endpoint runtime for MPI+OpenMP hybrid programming model on multi-core systems.},
  url = {http://dblp.uni-trier.de/db/conf/ppopp/ppopp2014.html#LuoLHKP14},
  series = {PPoPP '14},
  note = {Poster Paper},
  year = 2014
}

@inproceedings{conf-europar-ZhangLJSP14,
  added-at = {2018-04-22T00:00:00.000+0200},
  author = {Zhang, Jie and Lu, Xiaoyi and Jose, Jithin and Shi, Rong and Panda, Dhabaleswar K.},
  biburl = {https://www.bibsonomy.org/bibtex/2f1b8423c616caf01278e2ee1c569806d/dblp},
  booktitle = {Proceedings of International European Conference on Parallel Processing},
  crossref = {conf/europar/2014},
  editor = {Silva, Fernando M. A. and de Castro Dutra, Inês and Costa, Vítor Santos},
  ee = {https://doi.org/10.1007/978-3-319-09873-9_29},
  interhash = {dcf38e2cd6a02057abd5deb91837b1c7},
  intrahash = {f1b8423c616caf01278e2ee1c569806d},
  isbn = {978-3-319-09872-2},
  keywords = {dblp},
  pages = {342-353},
  publisher = {Springer},
  timestamp = {2019-10-22T13:26:26.000+0200},
  title = {Can Inter-VM Shmem Benefit MPI Applications on SR-IOV Based Virtualized Infiniband Clusters?},
  url = {http://dblp.uni-trier.de/db/conf/europar/europar2014.html#ZhangLJSP14},
  volume = 8632,
  series = {Euro-Par '14},
  year = 2014
}

@inproceedings{conf-hipc-ZhangLJLSP14,
  added-at = {2018-04-22T00:00:00.000+0200},
  author = {Zhang, Jie and Lu, Xiaoyi and Jose, Jithin and Li, Mingzhe and Shi, Rong and Panda, Dhabaleswar K.},
  biburl = {https://www.bibsonomy.org/bibtex/2eb647b2ce10826decf1ee5de99d4b586/dblp},
  booktitle = {Proceedings of IEEE International Conference on High Performance Computing},
  crossref = {conf/hipc/2014},
  ee = {http://doi.ieeecomputersociety.org/10.1109/HiPC.2014.7116876},
  interhash = {75e70727f590e48af288c8c5761d0461},
  intrahash = {eb647b2ce10826decf1ee5de99d4b586},
  isbn = {978-1-4799-5976-1},
  keywords = {dblp},
  pages = {1-10},
  publisher = {IEEE Computer Society},
  timestamp = {2019-10-17T16:00:49.000+0200},
  title = {High performance MPI library over SR-IOV enabled infiniband clusters.},
  url = {http://dblp.uni-trier.de/db/conf/hipc/hipc2014.html#ZhangLJLSP14},
  series = {HiPC '14},
  year = 2014
}

@inproceedings{conf-nas-LiangFLX14,
  added-at = {2020-04-08T00:00:00.000+0200},
  author = {Liang, Fan and Feng, Chen and Lu, Xiaoyi and Xu, Zhiwei},
  biburl = {https://www.bibsonomy.org/bibtex/2e3cf4d32e28bf14c02cf9e7056f95b73/dblp},
  booktitle = {Proceedings of IEEE International Conference on Networking, Architecture, and Storage},
  crossref = {conf/nas/2014},
  ee = {http://doi.ieeecomputersociety.org/10.1109/NAS.2014.39},
  interhash = {7a684c08f0aeb7f37ef33f5bf0d6585d},
  intrahash = {e3cf4d32e28bf14c02cf9e7056f95b73},
  isbn = {978-1-4799-4087-5},
  keywords = {dblp},
  pages = {207-215},
  publisher = {IEEE Computer Society},
  timestamp = {2020-04-09T11:43:01.000+0200},
  title = {Performance Characterization of Hadoop and Data MPI Based on Amdahl's Second Law.},
  url = {http://dblp.uni-trier.de/db/conf/nas/nas2014.html#LiangFLX14},
  series = {NAS '14},
  year = 2014
}

@inproceedings{conf-hoti-LuWISP14,
  added-at = {2015-08-13T00:00:00.000+0200},
  author = {Lu, Xiaoyi and ur Rahman, Md. Wasi and Islam, Nusrat S. and Shankar, Dipti and Panda, Dhabaleswar K.},
  biburl = {https://www.bibsonomy.org/bibtex/2111584e20430bfbf02620f4937580fe6/dblp},
  booktitle = {Proceedings of Annual Symposium on High-Performance Interconnects},
  crossref = {conf/hoti/2014},
  ee = {http://doi.ieeecomputersociety.org/10.1109/HOTI.2014.15},
  interhash = {853ff89efd6ec0135c8a21efd88677cc},
  intrahash = {111584e20430bfbf02620f4937580fe6},
  isbn = {978-1-4799-5860-3},
  keywords = {dblp},
  pages = {9-16},
  publisher = {IEEE Computer Society},
  timestamp = {2015-11-07T11:34:03.000+0100},
  title = {Accelerating Spark with RDMA for Big Data Processing: Early Experiences.},
  url = {http://dblp.uni-trier.de/db/conf/hoti/hoti2014.html#LuWISP14},
  series = {HotI '14},
  year = 2014
}

@inproceedings{conf-cluster-JoseHLPZTP14,
  added-at = {2018-04-22T00:00:00.000+0200},
  author = {Jose, Jithin and Hamidouche, Khaled and Lu, Xiaoyi and Potluri, Sreeram and Zhang, Jie and Tomko, Karen and Panda, Dhabaleswar K.},
  biburl = {https://www.bibsonomy.org/bibtex/2307a8463075a8f6cd3d8a7e033e8e662/dblp},
  booktitle = {Proceedings of IEEE International Conference on Cluster Computing},
  crossref = {conf/cluster/2014},
  ee = {http://doi.ieeecomputersociety.org/10.1109/CLUSTER.2014.6968754},
  interhash = {cbfa916a51844689848cbfb3c7aab960},
  intrahash = {307a8463075a8f6cd3d8a7e033e8e662},
  isbn = {978-1-4799-5548-0},
  keywords = {dblp},
  pages = {10-18},
  publisher = {IEEE Computer Society},
  timestamp = {2018-04-24T11:40:15.000+0200},
  title = {High performance OpenSHMEM for Xeon Phi clusters: Extensions, runtime designs and application co-design.},
  url = {http://dblp.uni-trier.de/db/conf/cluster/cluster2014.html#JoseHLPZTP14},
  series = {CLUSTER '14},
  highlight = {Best Paper Award Nomination},
  year = 2014
}

@inproceedings{conf-hpdc-IslamLWP14,
  added-at = {2018-11-06T00:00:00.000+0100},
  author = {Islam, Nusrat S. and Lu, Xiaoyi and ur Rahman, Md. Wasi and Panda, Dhabaleswar K.},
  biburl = {https://www.bibsonomy.org/bibtex/21c0c65aaa0d1b70bea685b7c5542e5a8/dblp},
  booktitle = {Proceedings of International ACM Symposium on High Performance and Distributed Computing},
  crossref = {conf/hpdc/2014},
  editor = {Plale, Beth and Ripeanu, Matei and Cappello, Franck and Xu, Dongyan},
  ee = {https://doi.org/10.1145/2600212.2600715},
  interhash = {b0527b250859d795e7041059a2f556e6},
  intrahash = {1c0c65aaa0d1b70bea685b7c5542e5a8},
  isbn = {978-1-4503-2749-7},
  keywords = {dblp},
  pages = {261-264},
  publisher = {ACM},
  timestamp = {2018-11-07T14:07:00.000+0100},
  title = {SOR-HDFS: a SEDA-based approach to maximize overlapping in RDMA-enhanced HDFS.},
  url = {http://dblp.uni-trier.de/db/conf/hpdc/hpdc2014.html#IslamLWP14},
  series = {HPDC '14},
  note = {Short Paper},
  year = 2014
}

@inproceedings{conf-icpp-ShiLPHZP14,
  added-at = {2018-04-22T00:00:00.000+0200},
  author = {Shi, Rong and Lu, Xiaoyi and Potluri, Sreeram and Hamidouche, Khaled and Zhang, Jie and Panda, Dhabaleswar K.},
  biburl = {https://www.bibsonomy.org/bibtex/2ff6dead7dc3b11cc85e18ac240ea12f2/dblp},
  booktitle = {Proceedings of International Conference on Parallel Processing},
  crossref = {conf/icpp/2014},
  ee = {http://doi.ieeecomputersociety.org/10.1109/ICPP.2014.31},
  interhash = {bc5d4b2381f0be53aea9ebd6568a8c6d},
  intrahash = {ff6dead7dc3b11cc85e18ac240ea12f2},
  keywords = {dblp},
  pages = {221-230},
  publisher = {IEEE Computer Society},
  timestamp = {2018-04-24T11:39:49.000+0200},
  title = {HAND: A Hybrid Approach to Accelerate Non-contiguous Data Movement Using MPI Datatypes on GPU Clusters.},
  url = {http://dblp.uni-trier.de/db/conf/icpp/icpp2014.html#ShiLPHZP14},
  series = {ICPP '14},
  year = 2014
}

@InProceedings{conf-bpoe-han14,
author="Han, Rui
and Lu, Xiaoyi
and Xu, Jiangtao",
title="On Big Data Benchmarking",
booktitle="Proceedings of Big Data Benchmarks, Performance Optimization, and Emerging Hardware",
year="2014",
publisher="Springer International Publishing",
address="Cham",
pages="3--18",
abstract="Big data systems address the challenges of capturing, storing, managing, analyzing, and visualizing big data. Within this context, developing benchmarks to evaluate and compare big data systems has become an active topic for both research and industry communities. To date, most of the state-of-the-art big data benchmarks are designed for specific types of systems. Based on our experience, however, we argue that considering the complexity, diversity, and rapid evolution of big data systems, for the sake of fairness, big data benchmarks must include diversity of data and workloads. Given this motivation, in this paper, we first propose the key requirements and challenges in developing big data benchmarks from the perspectives of generating data with 4 V properties (i.e. volume, velocity, variety and veracity) of big data, as well as generating tests with comprehensive workloads for big data systems. We then present the methodology on big data benchmarking designed to address these challenges. Next, the state-of-the-art are summarized and compared, following by our vision for future research directions.",
isbn="978-3-319-13021-7",
series = "BPOE '14"
}

@inproceedings{conf-vldb-LiangFLX14,
  added-at = {2020-04-08T00:00:00.000+0200},
  author = {Liang, Fan and Feng, Chen and Lu, Xiaoyi and Xu, Zhiwei},
  biburl = {https://www.bibsonomy.org/bibtex/26c52f87b6b582c2f9f223e1e33dfb575/dblp},
  booktitle = {Proceedings of International Workshop on Big Data Benchmarks, Performance Optimization, and Emerging Hardware},
  crossref = {conf/vldb/2014bpoe},
  editor = {Zhan, Jianfeng and Han, Rui and Weng, Chuliang},
  ee = {https://doi.org/10.1007/978-3-319-13021-7_9},
  interhash = {3a0193fc119c04e4d8b5f53132d93850},
  intrahash = {6c52f87b6b582c2f9f223e1e33dfb575},
  isbn = {978-3-319-13020-0},
  keywords = {dblp},
  pages = {111-123},
  publisher = {Springer},
  timestamp = {2020-04-09T11:43:10.000+0200},
  title = {Performance Benefits of DataMPI: A Case Study with BigDataBench.},
  url = {http://dblp.uni-trier.de/db/conf/vldb/bpoe2014.html#LiangFLX14},
  volume = 8807,
  series = {BPOE '14},
  year = 2014
}

@inproceedings{conf-icpp-Wasi-ur-RahmanLIP14,
  added-at = {2014-11-24T00:00:00.000+0100},
  author = {ur Rahman, Md. Wasi and Lu, Xiaoyi and Islam, Nusrat Sharmin and Panda, Dhabaleswar K.},
  biburl = {https://www.bibsonomy.org/bibtex/2e740f5fc2f9ba164d8d7a04f2d215efb/dblp},
  booktitle = {Proceedings of International Conference on Parallel Processing},
  crossref = {conf/icpp/2014},
  ee = {http://dx.doi.org/10.1109/ICPP.2014.14},
  interhash = {a337b29d5f5269b5acf5785dd33e7c5e},
  intrahash = {e740f5fc2f9ba164d8d7a04f2d215efb},
  keywords = {dblp},
  pages = {50-59},
  publisher = {IEEE Computer Society},
  timestamp = {2015-06-18T13:51:32.000+0200},
  title = {Performance Modeling for RDMA-Enhanced Hadoop MapReduce.},
  url = {http://dblp.uni-trier.de/db/conf/icpp/icpp2014.html#Wasi-ur-RahmanLIP14},
  series = {ICPP '14},
  year = 2014
}

@inproceedings{conf-cluster-LiLPHJTP14,
  added-at = {2015-11-05T00:00:00.000+0100},
  author = {Li, Mingzhe and Lu, Xiaoyi and Potluri, Sreeram and Hamidouche, Khaled and Jose, Jithin and Tomko, Karen and Panda, Dhabaleswar K.},
  biburl = {https://www.bibsonomy.org/bibtex/2be8ee1e530c7d28f6d8b994d2a759966/dblp},
  booktitle = {Proceedings of IEEE International Conference on Cluster Computing},
  crossref = {conf/cluster/2014},
  ee = {http://doi.ieeecomputersociety.org/10.1109/CLUSTER.2014.6968755},
  interhash = {71a17b919c866b5750f15af9045296b9},
  intrahash = {be8ee1e530c7d28f6d8b994d2a759966},
  isbn = {978-1-4799-5548-0},
  keywords = {dblp},
  pages = {230-238},
  publisher = {IEEE Computer Society},
  timestamp = {2015-11-06T11:35:12.000+0100},
  title = {Scalable Graph500 design with MPI-3 RMA.},
  url = {http://dblp.uni-trier.de/db/conf/cluster/cluster2014.html#LiLPHJTP14},
  series = {CLUSTER '14},
  year = 2014
}

@inproceedings{conf-europar-Wasi-ur-RahmanLIRP14,
  added-at = {2017-05-24T00:00:00.000+0200},
  author = {ur Rahman, Md. Wasi and Lu, Xiaoyi and Islam, Nusrat Sharmin and Rajachandrasekar, Raghunath and Panda, Dhabaleswar K.},
  biburl = {https://www.bibsonomy.org/bibtex/2a2eaf11da42a602b19587534ac94c5cc/dblp},
  booktitle = {Proceedings of International European Conference on Parallel Processing},
  crossref = {conf/europar/2014},
  editor = {Silva, Fernando M. A. and de Castro Dutra, Inês and Costa, Vítor Santos},
  ee = {https://doi.org/10.1007/978-3-319-09873-9_54},
  interhash = {ffbd4c417bd2621281357a4a2b17af6d},
  intrahash = {a2eaf11da42a602b19587534ac94c5cc},
  isbn = {978-3-319-09872-2},
  keywords = {dblp},
  pages = {644-655},
  publisher = {Springer},
  timestamp = {2019-10-22T13:26:26.000+0200},
  title = {MapReduce over Lustre: Can RDMA-Based Approach Benefit?},
  url = {http://dblp.uni-trier.de/db/conf/europar/europar2014.html#Wasi-ur-RahmanLIRP14},
  volume = 8632,
  series = {Euro-Par '14},
  year = 2014
}

@inproceedings{conf-ipps-LuLWZX14,
  added-at = {2020-04-08T00:00:00.000+0200},
  author = {Lu, Xiaoyi and Liang, Fan and Wang, Bing and Zha, Li and Xu, Zhiwei},
  biburl = {https://www.bibsonomy.org/bibtex/2c8573723bd8a72ff2ab5df6212b3c739/dblp},
  booktitle = {Proceedings of IEEE International Parallel and Distributed Processing Symposium},
  crossref = {conf/ipps/2014},
  ee = {http://doi.ieeecomputersociety.org/10.1109/IPDPS.2014.90},
  interhash = {b7b5263d58f96ddd0ce9d482aa71092e},
  intrahash = {c8573723bd8a72ff2ab5df6212b3c739},
  isbn = {978-1-4799-3799-8},
  keywords = {dblp},
  pages = {829-838},
  publisher = {IEEE Computer Society},
  timestamp = {2020-04-09T11:42:17.000+0200},
  title = {DataMPI: Extending MPI to Hadoop-Like Big Data Computing.},
  url = {http://dblp.uni-trier.de/db/conf/ipps/ipdps2014.html#LuLWZX14},
  series = {IPDPS '14},
  year = 2014
}

@inproceedings{conf-ics-Wasi-ur-RahmanLIP14,
  added-at = {2018-11-06T00:00:00.000+0100},
  author = {ur Rahman, Md. Wasi and Lu, Xiaoyi and Islam, Nusrat Sharmin and Panda, Dhabaleswar K.},
  biburl = {https://www.bibsonomy.org/bibtex/24a0a826741f3172cfcd8942072de5a4a/dblp},
  booktitle = {Proceedings of International Conference on Supercompuing},
  crossref = {conf/ics/2014},
  editor = {Bode, Arndt and Gerndt, Michael and Stenström, Per and Rauchwerger, Lawrence and Miller, Barton P. and Schulz, Martin},
  ee = {https://doi.org/10.1145/2597652.2597684},
  interhash = {21285a5d648b653c1f46692035218486},
  intrahash = {4a0a826741f3172cfcd8942072de5a4a},
  isbn = {978-1-4503-2642-1},
  keywords = {dblp},
  pages = {33-42},
  publisher = {ACM},
  timestamp = {2018-11-07T14:15:50.000+0100},
  title = {HOMR: a hybrid approach to exploit maximum overlapping in MapReduce over high performance interconnects.},
  url = {http://dblp.uni-trier.de/db/conf/ics/ics2014.html#Wasi-ur-RahmanLIP14},
  series = {ICS '14},
  year = 2014
}

@inproceedings{conf-vldb-ShankarLWIP14,
  added-at = {2017-05-25T00:00:00.000+0200},
  author = {Shankar, Dipti and Lu, Xiaoyi and ur Rahman, Md. Wasi and Islam, Nusrat S. and Panda, Dhabaleswar K.},
  biburl = {https://www.bibsonomy.org/bibtex/27402889c809665d12a6f97370064364e/dblp},
  booktitle = {Proceedings of International Workshop on Big Data Benchmarks, Performance Optimization, and Emerging Hardware},
  crossref = {conf/vldb/2014bpoe},
  editor = {Zhan, Jianfeng and Han, Rui and Weng, Chuliang},
  ee = {https://doi.org/10.1007/978-3-319-13021-7_2},
  interhash = {119a585b1ecc22abcfd866d5c89bb059},
  intrahash = {7402889c809665d12a6f97370064364e},
  isbn = {978-3-319-13020-0},
  keywords = {dblp},
  pages = {19-33},
  publisher = {Springer},
  timestamp = {2019-05-15T14:22:23.000+0200},
  title = {A Micro-benchmark Suite for Evaluating Hadoop MapReduce on High-Performance Networks.},
  url = {http://dblp.uni-trier.de/db/conf/vldb/bpoe2014.html#ShankarLWIP14},
  volume = 8807,
  series = {BPOE '14},
  year = 2014
}

@inproceedings{conf-pgas-JosePSLHSSP14,
  added-at = {2018-11-06T00:00:00.000+0100},
  author = {Jose, Jithin and Potluri, Sreeram and Subramoni, Hari and Lu, Xiaoyi and Hamidouche, Khaled and Schulz, Karl W. and Sundar, Hari and Panda, Dhabaleswar K.},
  biburl = {https://www.bibsonomy.org/bibtex/2fc6eebebb56250093d464ebb7b4b89c5/dblp},
  booktitle = {Proceedings of International Conference on Partitioned Global Address Space Programming Models},
  crossref = {conf/pgas/2014},
  editor = {Malony, Allen D. and Hammond, Jeff R.},
  ee = {https://doi.org/10.1145/2676870.2676880},
  interhash = {016b99735122bfb3e7f9e4dba9363ef6},
  intrahash = {fc6eebebb56250093d464ebb7b4b89c5},
  isbn = {978-1-4503-3247-7},
  keywords = {dblp},
  pages = {7:1-7:9},
  publisher = {ACM},
  timestamp = {2018-11-07T14:54:20.000+0100},
  title = {Designing Scalable Out-of-core Sorting with Hybrid MPI+PGAS Programming Models.},
  url = {http://dblp.uni-trier.de/db/conf/pgas/pgas2014.html#JosePSLHSSP14},
  series = {PGAS '14},
  year = 2014
}

@inproceedings{conf-wbdb-LuWIP13,
  added-at = {2017-05-21T00:00:00.000+0200},
  author = {Lu, Xiaoyi and ur Rahman, Md. Wasi and Islam, Nusrat Sharmin and Panda, Dhabaleswar K.},
  biburl = {https://www.bibsonomy.org/bibtex/27fa5f1449ebe954e638500ad75c435be/dblp},
  booktitle = {Proceedings of the 3rd Workshop on Big Data Benchmarking},
  crossref = {conf/wbdb/2013},
  editor = {Rabl, Tilmann and Jacobsen, Hans-Arno and Nambiar, Raghunath and Poess, Meikel and Bhandarkar, Milind A. and Baru, Chaitanya K.},
  ee = {https://doi.org/10.1007/978-3-319-10596-3_3},
  interhash = {f44761b2ce8a2f84771dcf0db9421e4f},
  intrahash = {7fa5f1449ebe954e638500ad75c435be},
  isbn = {978-3-319-10595-6},
  keywords = {dblp},
  pages = {32-42},
  publisher = {Springer},
  timestamp = {2019-05-15T13:01:26.000+0200},
  title = {A Micro-benchmark Suite for Evaluating Hadoop RPC on High-Performance Networks.},
  url = {http://dblp.uni-trier.de/db/conf/wbdb/wbdb2013.html#LuWIP13},
  volume = 8585,
  series = {WBDB '13},
  year = 2013
}

@inproceedings{conf-ipps-Wasi-ur-RahmanILJSWP13,
  added-at = {2017-05-24T00:00:00.000+0200},
  author = {ur Rahman, Md. Wasi and Islam, Nusrat Sharmin and Lu, Xiaoyi and Jose, Jithin and Subramoni, Hari and Wang, Hao and Panda, Dhabaleswar K.},
  biburl = {https://www.bibsonomy.org/bibtex/2a3e0be2905f56c79cfb2cd32f947d6fe/dblp},
  booktitle = {Workshop Proceedings of IEEE International Parallel and Distributed Processing Symposium},
  crossref = {conf/ipps/2013w},
  ee = {http://doi.ieeecomputersociety.org/10.1109/IPDPSW.2013.238},
  interhash = {dc1c25ed4f59ef5fe4775d3a2f54f145},
  intrahash = {a3e0be2905f56c79cfb2cd32f947d6fe},
  keywords = {dblp},
  pages = {1908-1917},
  publisher = {IEEE},
  timestamp = {2019-10-17T20:35:00.000+0200},
  title = {High-Performance RDMA-based Design of Hadoop MapReduce over InfiniBand.},
  url = {http://dblp.uni-trier.de/db/conf/ipps/ipdps2013w.html#Wasi-ur-RahmanILJSWP13},
  series = {IPDPS Workshops '13},
  year = 2013
}

@inproceedings{conf-cloud-Wasi-ur-RahmanLIP13,
  added-at = {2018-11-06T00:00:00.000+0100},
  author = {ur Rahman, Md. Wasi and Lu, Xiaoyi and Islam, Nusrat S. and Panda, Dhabaleswar K.},
  biburl = {https://www.bibsonomy.org/bibtex/2abd0475717d2032a49e656b54a80dd49/dblp},
  booktitle = {Proceedings of ACM Symposium on Cloud Computing},
  crossref = {conf/cloud/2013},
  editor = {Lohman, Guy M.},
  ee = {https://doi.org/10.1145/2523616.2525953},
  interhash = {e791b903a0c4142d148f8df944a7a2a5},
  intrahash = {abd0475717d2032a49e656b54a80dd49},
  isbn = {978-1-4503-2428-1},
  keywords = {dblp},
  pages = {45:1-45:2},
  publisher = {ACM},
  timestamp = {2018-11-07T16:53:11.000+0100},
  title = {Does RDMA-based enhanced Hadoop MapReduce need a new performance model?},
  url = {http://dblp.uni-trier.de/db/conf/cloud/socc2013.html#Wasi-ur-RahmanLIP13},
  series = {SoCC '13},
  note = {Poster Paper},
  year = 2013
}

@inproceedings{conf-hoti-IslamLWP13,
  added-at = {2014-12-02T00:00:00.000+0100},
  author = {Islam, Nusrat S. and Lu, Xiaoyi and ur Rahman, Md. Wasi and Panda, Dhabaleswar K.},
  biburl = {https://www.bibsonomy.org/bibtex/21af288e6acbf8fff2f51d1befbba7083/dblp},
  booktitle = {Proceedings of Annual Symposium on High-Performance Interconnects},
  crossref = {conf/hoti/2013},
  ee = {http://doi.ieeecomputersociety.org/10.1109/HOTI.2013.24},
  interhash = {1ff5395206a7e9dd6ce18653617af521},
  intrahash = {1af288e6acbf8fff2f51d1befbba7083},
  isbn = {978-0-7695-5103-6},
  keywords = {dblp},
  pages = {75-78},
  publisher = {IEEE Computer Society},
  timestamp = {2015-06-18T12:10:31.000+0200},
  title = {Can Parallel Replication Benefit Hadoop Distributed File System for High Performance Interconnects?},
  url = {http://dblp.uni-trier.de/db/conf/hoti/hoti2013.html#IslamLWP13},
  series = {HotI '13},
  note = {Short Paper},
  year = 2013
}

@inproceedings{conf-icpp-LuIWJSWP13,
  added-at = {2016-07-14T00:00:00.000+0200},
  author = {Lu, Xiaoyi and Islam, Nusrat S. and ur Rahman, Md. Wasi and Jose, Jithin and Subramoni, Hari and Wang, Hao and Panda, Dhabaleswar K.},
  biburl = {https://www.bibsonomy.org/bibtex/2ac7bbfd647600c13b5621a0d56f211b2/dblp},
  booktitle = {Proceedings of International Conference on Parallel Processing},
  crossref = {conf/icpp/2013},
  ee = {http://doi.ieeecomputersociety.org/10.1109/ICPP.2013.78},
  interhash = {78a275bba85f569ef1b5bd8538cbea23},
  intrahash = {ac7bbfd647600c13b5621a0d56f211b2},
  isbn = {978-0-7695-5117-3},
  keywords = {dblp},
  pages = {641-650},
  publisher = {IEEE Computer Society},
  timestamp = {2016-07-15T11:40:31.000+0200},
  title = {High-Performance Design of Hadoop RPC with RDMA over InfiniBand.},
  url = {http://dblp.uni-trier.de/db/conf/icpp/icpp2013.html#LuIWJSWP13},
  series = {ICPP '13},
  year = 2013
}

@inproceedings{conf-ccgrid-JoseLLKAP13,
  added-at = {2013-06-27T00:00:00.000+0200},
  author = {Jose, Jithin and Li, Mingzhe and Lu, Xiaoyi and Kandalla, Krishna Chaitanya and Arnold, Mark Daniel and Panda, Dhabaleswar K.},
  biburl = {https://www.bibsonomy.org/bibtex/2052a0c90ed4307de38c2cf6843b56582/dblp},
  booktitle = {Proceedings of IEEE/ACM International Symposium on Cluster, Cloud and Grid Computing},
  crossref = {conf/ccgrid/2013},
  ee = {http://doi.ieeecomputersociety.org/10.1109/CCGrid.2013.76},
  interhash = {8794ef52b16a38e38183855ea808cad9},
  intrahash = {052a0c90ed4307de38c2cf6843b56582},
  isbn = {978-1-4673-6465-2},
  keywords = {dblp},
  pages = {385-392},
  publisher = {IEEE Computer Society},
  timestamp = {2016-01-20T11:43:19.000+0100},
  title = {SR-IOV Support for Virtualization on InfiniBand Clusters: Early Experience.},
  url = {http://dblp.uni-trier.de/db/conf/ccgrid/ccgrid2013.html#JoseLLKAP13},
  series = {CCGRID '13},
  highlight = {Best Presentation Award Nomination},
  year = 2013
}

@inproceedings{conf-cluster-ShiPHLTP13,
  added-at = {2014-11-04T00:00:00.000+0100},
  author = {Shi, Rong and Potluri, Sreeram and Hamidouche, Khaled and Lu, Xiaoyi and Tomko, Karen and Panda, Dhabaleswar K.},
  biburl = {https://www.bibsonomy.org/bibtex/2b909cf61ac17bcbdeef802fe12534587/dblp},
  booktitle = {Proceedings of IEEE International Conference on Cluster Computing},
  crossref = {conf/cluster/2013},
  ee = {http://doi.ieeecomputersociety.org/10.1109/CLUSTER.2013.6702619},
  interhash = {a007882a649dd5cae714900b7716dcf3},
  intrahash = {b909cf61ac17bcbdeef802fe12534587},
  isbn = {978-1-4799-0898-1},
  keywords = {dblp},
  pages = {1-8},
  publisher = {IEEE Computer Society},
  timestamp = {2015-06-18T13:01:09.000+0200},
  title = {A scalable and portable approach to accelerate hybrid HPL on heterogeneous CPU-GPU clusters.},
  url = {http://dblp.uni-trier.de/db/conf/cluster/cluster2013.html#ShiPHLTP13},
  series = {CLUSTER '13},
  highlight = {Best Student Paper Award},
  year = 2013
}

@inproceedings{conf-wbdb-IslamLWJP12,
  added-at = {2017-05-21T00:00:00.000+0200},
  author = {Islam, Nusrat Sharmin and Lu, Xiaoyi and ur Rahman, Md. Wasi and Jose, Jithin and Panda, Dhabaleswar K.},
  biburl = {https://www.bibsonomy.org/bibtex/28fcd773251a852c2e709510ad0a140c3/dblp},
  booktitle = {Proceedings of the 3rd Workshop on Big Data Benchmarking},
  crossref = {conf/wbdb/2012},
  editor = {Rabl, Tilmann and Poess, Meikel and Baru, Chaitanya K. and Jacobsen, Hans-Arno},
  ee = {https://doi.org/10.1007/978-3-642-53974-9_12},
  interhash = {0772224f15564db8227c3307a765628f},
  intrahash = {8fcd773251a852c2e709510ad0a140c3},
  isbn = {978-3-642-53973-2},
  keywords = {dblp},
  pages = {129-147},
  publisher = {Springer},
  timestamp = {2019-05-15T13:01:26.000+0200},
  title = {A Micro-benchmark Suite for Evaluating HDFS Operations on Modern Clusters.},
  url = {http://dblp.uni-trier.de/db/conf/wbdb/wbdb2012.html#IslamLWJP12},
  volume = 8163,
  series = {WBDB '12},
  year = 2012
}

@inproceedings{conf-ispa-LuLZX11,
  added-at = {2020-04-08T00:00:00.000+0200},
  author = {Lu, Xiaoyi and Lin, Jian and Zha, Li and Xu, Zhiwei},
  biburl = {https://www.bibsonomy.org/bibtex/2f6594644bfa295ebc7ff0f85cec472d3/dblp},
  booktitle = {Proceedings of IEEE International Symposium on Parallel and Distributed Processing with Applications},
  crossref = {conf/ispa/2011},
  ee = {http://doi.ieeecomputersociety.org/10.1109/ISPA.2011.58},
  interhash = {be752a8362cf72700e0f9d3a15a23d24},
  intrahash = {f6594644bfa295ebc7ff0f85cec472d3},
  isbn = {978-1-4577-0391-1},
  keywords = {dblp},
  pages = {99-106},
  publisher = {IEEE Computer Society},
  timestamp = {2020-04-09T11:43:03.000+0200},
  title = {Vega LingCloud: A Resource Single Leasing Point System to Support Heterogeneous Application Modes on Shared Infrastructure.},
  url = {http://dblp.uni-trier.de/db/conf/ispa/ispa2011.html#LuLZX11},
  series = {ISPA '11},
  highlight = {Best Paper Award},
  year = 2011
}

@inproceedings{conf-icppw-LuWZX11,
  added-at = {2020-04-08T00:00:00.000+0200},
  author = {Lu, Xiaoyi and Wang, Bing and Zha, Li and Xu, Zhiwei},
  biburl = {https://www.bibsonomy.org/bibtex/2a176fb601f0046f1c1895834f75ea5fb/dblp},
  booktitle = {Workshop Proceedings of International Conference on Parallel Processing},
  crossref = {conf/icppw/2011},
  editor = {Sheu, Jang-Ping and Wang, Cho-Li},
  ee = {http://doi.ieeecomputersociety.org/10.1109/ICPPW.2011.56},
  interhash = {0d17e498248d559639a6a085e6869b9d},
  intrahash = {a176fb601f0046f1c1895834f75ea5fb},
  isbn = {978-1-4577-1337-8},
  keywords = {dblp},
  pages = {371-379},
  publisher = {IEEE Computer Society},
  timestamp = {2020-04-09T11:41:44.000+0200},
  title = {Can MPI Benefit Hadoop and MapReduce Applications?},
  url = {http://dblp.uni-trier.de/db/conf/icppw/icppw2011.html#LuWZX11},
  series = {ICPP Workshops '11},
  year = 2011
}

@inproceedings{conf-nas-LinLYZZ10,
  added-at = {2020-09-05T00:00:00.000+0200},
  author = {Lin, Jian and Lu, Xiaoyi and Yu, Lin and Zou, Yongqiang and Zha, Li},
  biburl = {https://www.bibsonomy.org/bibtex/244e2147962a29168a51368ae80ee4271/dblp},
  booktitle = {Proceedings of IEEE International Conference on Networking, Architecture, and Storage},
  crossref = {conf/nas/2010},
  ee = {http://doi.ieeecomputersociety.org/10.1109/NAS.2010.34},
  interhash = {05abb8f399dd86735e3a69071acafab3},
  intrahash = {44e2147962a29168a51368ae80ee4271},
  isbn = {978-0-7695-4134-1},
  keywords = {dblp},
  pages = {457-464},
  publisher = {IEEE Computer Society},
  timestamp = {2020-09-09T14:59:33.000+0200},
  title = {VegaWarden: A Uniform User Management System for Cloud Applications.},
  url = {http://dblp.uni-trier.de/db/conf/nas/nas2010.html#LinLYZZ10},
  series = {NAS '10},
  year = 2010
}

@inproceedings{conf-services-LuLZPLZ10,
  added-at = {2017-05-23T00:00:00.000+0200},
  author = {Lu, Xiaoyi and Lin, Jian and Zou, Yongqiang and Peng, Juan and Liu, Xingwu and Zha, Li},
  biburl = {https://www.bibsonomy.org/bibtex/20e8b27b67c37c2224fab84bd3e04555b/dblp},
  booktitle = {Proceedings of World Congress on Services},
  crossref = {conf/services/2010},
  ee = {http://doi.ieeecomputersociety.org/10.1109/SERVICES.2010.58},
  interhash = {f69f2771fa330e329300cff37e094460},
  intrahash = {0e8b27b67c37c2224fab84bd3e04555b},
  isbn = {978-0-7695-4129-7},
  keywords = {dblp},
  pages = {375-382},
  publisher = {IEEE Computer Society},
  timestamp = {2019-10-17T15:33:50.000+0200},
  title = {Investigating, Modeling, and Ranking Interface Complexity of Web Services on the World Wide Web.},
  url = {http://dblp.uni-trier.de/db/conf/services/services2010.html#LuLZPLZ10},
  series = {SERVICES '10},
  year = 2010
}

@inproceedings{conf-npc-PengLCZ10,
  added-at = {2018-06-26T00:00:00.000+0200},
  author = {Peng, Juan and Lu, Xiaoyi and Cheng, Boqun and Zha, Li},
  biburl = {https://www.bibsonomy.org/bibtex/20fe8536307a51be116e4f002f7e948f4/dblp},
  booktitle = {Proceedings of IFIP International Conference on Network and Parallel Computing},
  crossref = {conf/npc/2010},
  editor = {Ding, Chen and Shao, Zhiyuan and Zheng, Ran},
  ee = {https://doi.org/10.1007/978-3-642-15672-4_34},
  interhash = {8c0766f1c8b10ba14f2a0ebb5cb8587c},
  intrahash = {0fe8536307a51be116e4f002f7e948f4},
  isbn = {978-3-642-15671-7},
  keywords = {dblp},
  pages = {412-422},
  publisher = {Springer},
  timestamp = {2019-05-15T14:23:05.000+0200},
  title = {JAMILA: A Usable Batch Job Management System to Coordinate Heterogeneous Clusters and Diverse Applications over Grid or Cloud Infrastructure.},
  url = {http://dblp.uni-trier.de/db/conf/npc/npc2010.html#PengLCZ10},
  volume = 6289,
  series = {NPC '10},
  year = 2010
}

@inproceedings{conf-services-YueLSXYZ09,
  added-at = {2020-04-08T00:00:00.000+0200},
  author = {Yue, Qiang and Lu, Xiaoyi and Shan, Zhiguang and Xu, Zhiwei and Yu, Haiyan and Zha, Li},
  biburl = {https://www.bibsonomy.org/bibtex/28aa099cee0c1814c0805b2ed331f7e31/dblp},
  booktitle = {Proceedings of World Congress on Services},
  crossref = {conf/services/2009-1},
  ee = {http://doi.ieeecomputersociety.org/10.1109/SERVICES-I.2009.20},
  interhash = {d68424d3725de74d6725c31c71f99dfd},
  intrahash = {8aa099cee0c1814c0805b2ed331f7e31},
  isbn = {978-0-7695-3708-5},
  keywords = {dblp},
  pages = {155-162},
  publisher = {IEEE Computer Society},
  timestamp = {2020-04-09T11:42:32.000+0200},
  title = {A Model of Message-Based Debugging Facilities for Web or Grid Services.},
  url = {http://dblp.uni-trier.de/db/conf/services/services2009-1.html#YueLSXYZ09},
  series = {SERVICES '09},
  year = 2009
}

@inproceedings{conf-pdcat-LuZXLZ09,
  added-at = {2015-03-26T00:00:00.000+0100},
  author = {Lu, Xiaoyi and Zou, Yongqiang and Xiong, Fei and Lin, Jian and Zha, Li},
  biburl = {https://www.bibsonomy.org/bibtex/230591aa533fd2348b6dd6475e21b02db/dblp},
  booktitle = {Proceedings of International Conference on Parallel and Distributed Computing, Applications, and Technologies},
  crossref = {conf/pdcat/2009},
  ee = {http://doi.ieeecomputersociety.org/10.1109/PDCAT.2009.74},
  interhash = {70f7f12cf08bc458dd7986c4da618503},
  intrahash = {30591aa533fd2348b6dd6475e21b02db},
  isbn = {978-0-7695-3914-0},
  keywords = {dblp},
  pages = {186-194},
  publisher = {IEEE Computer Society},
  timestamp = {2015-06-18T14:17:41.000+0200},
  title = {ICOMC: Invocation Complexity Of Multi-Language Clients for Classified Web Services and its Impact on Large Scale SOA Applications.},
  url = {http://dblp.uni-trier.de/db/conf/pdcat/pdcat2009.html#LuZXLZ09},
  series = {PDCAT '09},
  year = 2009
}

@inproceedings{conf-pdcat-LuYZW08,
  added-at = {2015-03-26T00:00:00.000+0100},
  author = {Lu, Xiaoyi and Yue, Qiang and Zou, Yongqiang and Wang, Xiaoning},
  biburl = {https://www.bibsonomy.org/bibtex/2d0a39af0a019593880f56ab39e74cd00/dblp},
  booktitle = {Proceedings of International Conference on Parallel and Distributed Computing, Applications, and Technologies},
  crossref = {conf/pdcat/2008},
  ee = {http://doi.ieeecomputersociety.org/10.1109/PDCAT.2008.30},
  interhash = {a58c9ca4ac41137975dd82153057f30c},
  intrahash = {d0a39af0a019593880f56ab39e74cd00},
  isbn = {978-0-7695-3443-5},
  keywords = {dblp},
  pages = {33-36},
  publisher = {IEEE Computer Society},
  timestamp = {2015-06-18T14:18:07.000+0200},
  title = {An Experimental Analysis for Memory Usage of GOS Core.},
  url = {http://dblp.uni-trier.de/db/conf/pdcat/pdcat2008.html#LuYZW08},
  series = {PDCAT '08},
  note = {Short Paper},
  year = 2008
}