Kento Sato


Email: kento@llnl.gov
Phone: +19254226918


Kento Sato is a postdoctoral researcher in the  Center for Applied Scientific Computing at Lawrence Livermore National Laboratory (LLNL). His research area is distributed systems and parallel computing, particularly in High Performance Computing (HPC). Major focuses of his research are application reproducibility (MPI reproducibility, and Validation), scalable fault tolerance (Scalable checkpoint/restart, Fault tolerant MPI, Resilient system design), and I/O optimization (NVRAM, Burst buffer, Big data), and cloud computing. He is also a member of the Scalable Checkpoint/Restart project team at LLNL.

He received his Ph.D. in the Dept. of Mathematical & Computing Sciences at Tokyo Tech in 2014, his M.S. in the Dept. of Mathematical & Computing Sciences at Tokyo Tech in 2010, and his B.S. in the Dept. of Information Science at Tokyo Tech in 2008.


Keywords

  • Fault tolerance
  • Debugging tools
  • MPI and Communication
  • I/O, Burst buffers and Data compression
  • Cloud computing
  • Others: GPU, DSL etc.

Personal website

kento.github.io
This website is managed by Kento Sato, and is not affiliated with or supported by LLNL, or any other organizations.

Conferences

  1. Kento Sato, Dong H. Ahn, Ignacio Laguna, Gregory L. Lee, Martin Schulz and Christopher M. Chambreau, “Noise Injection Techniques for Reproducing Subtle and Unintended Message Races”, Proceedings of the 20th ACM SIGPLAN Symposium on Principles and Practice of Parallel Programming (PPoPP17), Austin, USA, Feb, 2017.

  2. Tianqi Xu, Kento Sato and Satoshi Matsuoka, "CloudBB: Scalable I/O Accelerator for Shared Cloud Storage", The 22nd IEEE International Conference on Parallel and Distributed Systems (ICPADS 2016), Wuhan, China, Dec, 2016.

  3. Teng Wang, Kathryn Mohror, Adam Moody, Kento Sato and Weikuan Yu "An Ephemeral Burst-Buffer File System for Scientific Applications", In Proceedings of the International Conference on High Performance Computing, Networking, Storage and Analysis 2016 (SC16), Salt Lake City, USA, Nov, 2016. (Acceptance rate: 18.4%, 82/446) 
  4. Kento Sato, Dong H. Ahn, Ignacio Laguna, Gregory L. Lee and Martin Schulz, "Clock Delta Compression for Scalable Order-Replay of Non-Deterministic Parallel Applications", In Proceedings of the International Conference on High Performance Computing, Networking, Storage and Analysis 2015 (SC15), Austin, USA, Nov, 2015. (Acceptance rate: 22.1%, 79/358) 
  5. Naoto Sasaki, Kento Sato, Toshio Endo and Satoshi Matsuoka, "Exploration of Lossy Compression for Application-level Checkpoint/Restart", In Proceedings of the International Conference on Parallel and Distributed Processing Symposium 2015 (IPDPS2015), Hyderabad, INDIA, May, 2015. (Acceptance rate: 21.8%, 108/496) 
  6. Kento Sato, Kathryn Mohror, Adam Moody, Todd Gamblin, Bronis R. de Supinski, Naoya Maruyama and Satoshi Matsuoka, "A User-level InfiniBand-based File System and Checkpoint Strategy for Burst Buffers", In Proceedings of the 14th IEEE/ACM International Symposium on Cluster, Cloud and Grid Computing (CCGrid2014), Chicago, USA, May, 2014. (Acceptance rate: 19.1%, 54/283) (Best Paper Award, 1/54) 
  7. Kento Sato, Adam Moody, Kathryn Mohror, Todd Gamblin, Bronis R. de Supinski, Naoya Maruyama and Satoshi Matsuoka, "FMI: Fault Tolerant Messaging Interface for Fast and Transparent Recovery", In Proceedings of the International Conference on Parallel and Distributed Processing Symposium 2014 (IPDPS2014), Phoenix, USA, May, 2014. (Acceptance rate: 21.1%, 114/541) 
  8. Kento Sato, Adam Moody, Kathryn Mohror, Todd Gamblin, Bronis R. de Supinski, Naoya Maruyama and Satoshi Matsuoka, "Design and Modeling of a Non-blocking Checkpointing System", In Proceedings of the International Conference on High Performance Computing, Networking, Storage and Analysis 2012 (SC12), Salt Lake, USA, Nov, 2012. (Acceptance rate: 21.2%, 100/472) 
  9. Akira Nukada, Kento Sato and Satoshi Matsuoka, "Scalable Multi-GPU 3-D FFT for TSUBAME 2.0 Supercomputer", In Proceedings of the International Conference on High Performance Computing, Networking, Storage and Analysis 2012 (SC12), Salt Lake, USA, Nov, 2012. (Acceptance rate: 21.2%, 100/472) 
  10. Naoya Maruyama, Tatsuo Nomura, Kento Sato, and Satoshi Matsuoka, "Physis: An Implicitly Parallel Programming Model for Stencil Computations on Large-Scale GPU-Accelerated Supercomputers", In Proceedings of the International Conference on High Performance Computing, Networking, Storage and Analysis 2011 (SC11), Seattle, USA, Nov 2011. (Acceptance rate: 21.0%, 74/352) 

Workshops (Refereed)

  1. Aiman Fang, Ignacio Laguna, Kento Sato, Tanzima Islam and Kathryn Mohror, “Fault Tolerance Assistant (FTA): An Exception Handling Approach for MPI Programs” (Hote topic), Workshop on Exascale MPI (ExaMPI15) at Supercomputing 2015 (SC15), Austin, USA, Nov, 2015.
  2. Moshe Gabel, Kento Sato, Daniel Keren, Satoshi Matsuoka and Assaf Schuster, “Latent Fault Detection With Unbalanced Workloads”, Event Processing, Forecast- ing and Decision-Making in the Big Data Era 2015 (EPForDM2015) in conjunc- tion with the 18th International Conference on Extending Database Technology (EDBT), Brussels, Belgium, March, 2015.
  3. Takafumi Saito, Kento Sato, Hitoshi Sato and Satoshi Matsuoka, "Energy-aware I/O Optimization for Checkpoint and Restart on a NAND Flash Memory System", In the Workshop on Fault-Tolerance for HPC at Extreme Scale 2013 (FTXS2013) in conjunction with the International Symposium on High Performance Parallel and Distributed Computing (HPDC13), New York, USA, June, 2013.
  4. Kento Sato, Hitoshi Sato and Satoshi Matsuoka, "A Model-Based Algorithm for Optimizing I/O Intensive Applications in Clouds using VM-Based Migration", In the International Workshop On Cloud Computing (Cloud2009) in conjunction with the International Symposium on Cluster Computing and the Grid 2009 (CCGrid2009), Shanghai, China, May, 2009.
  5. 佐藤賢斗, 佐藤仁, 松岡聡, "仮想マシン動的再配置による大規模データアクセスの高速化", 情報処理学会先進的計算基盤システムシンポジウム論文集 (SACSIS2010), 奈良, May, 2010.

Workshops (Non-refereed)

  1. Tianqi Xu, Kento Sato and Satoshi Matsuoka, "Cloud-based Burst Buffers for I/O Acceleration", IPSJ SIG Technical Reports 2015-HPC-150, Beppu, Japan, Aug, 2015.
  2. Tianqi Xu, Kento Sato and Satoshi Matsuoka, "Towards Cloud Bursting for Extreme Scale Supercomputers", IPSJ SIG Technical Reports 2014-HPC-145, Nigata, Japan, Jul, 2014.
  3. 佐々木尚人, 佐藤賢斗, 遠藤敏夫, 松岡聡, "実アプリケーションにおけるウェーブレット変換を用いたチェックポイントデータの非可逆圧縮手法", IPSJ SIG Technical Reports 2014-HPC-145, Nigata, Japan, Jul, 2014
  4. Kento Sato, Satoshi Matsuoka, Adam Moody, Kathryn Mohror, Todd Gamblin, Bronis R. de Supinski and Naoya Maruyama, "Burst SSD Buffer: Checkpoint Strategy at Extreme Scale", IPSJ SIG Technical Reports 2013-HPC-141, Okinawa, Japan, Sep, 2013.
  5. 松岡 聡, 佐藤 賢斗, 遠藤敏夫 "エクサスケールスパコンに向けた耐故障性の評価 --- TSUBAME2.0を例にして ---", IPSJ SIG Technical Reports 2013-HPC-141, Okinawa, Japan, Sep, 2013.
  6. Kento Sato, Adam Moody, Kathryn Mohror, Todd Gamblin, Bronis R. de Supinski, Naoya Maruyama and Satoshi Matsuoka, "Design and Modeling of an Asynchronous Checkpointing System", IPSJ SIG Technical Reports 2012-HPC-135 (SWoPP 2012), Tottori, Aug, 2012.
  7. Kento Sato, Adam Moody, Kathryn Mohror, Todd Gamblin, Bronis R. de Supinski, Naoya Maruyama and Satoshi Matsuoka, "Towards an Asynchronous Checkpointing System", IPSJ SIG Technical Reports 2011-ARC-197 2011-HPC-132 (HOKKE-19), Hokkaido, Nov, 2011.
  8. 佐藤賢斗, 佐藤仁, 松岡聡, "仮想クラスタを用いたData-Intensive Application実行環境の性能モデル構築と最適化", 並列/分散/協調処理に関するサマーワークショップ (SWoPP2008), Saga, Aug, 2008.

Posters (Refereed)

  1. Teng Wang, Kathryn Mohror, Adam Moody, Weikuan Yu and Kento Sato, “BurstFS: A Distributed Burst Buffer File System for Scientific Applications”, In Proceedings of the International Conference on High Performance Computing, Networking, Storage and Analysis 2015 (SC15), Austin, USA, Nov, 2015. (Acceptance rate: 44%, 112/253)
  2. Tianqi Xu, Kento Sato and Satoshi Matsuoka, “Design and Modelling of Cloud-based Burst Buffers”, In Proceedings of the International Conference on High Performance Computing, Networking, Storage and Analysis 2015 (SC15), Austin, USA, Nov, 2015. (Acceptance rate: 44%, 112/253)
  3. Tianqi Xu, Kento Sato and Satoshi Matsuoka, “Towards Cloud-based Burst Buffers for I/O Intensive Computing in Cloud”, In HPC in Asia Workshop in conjunction with the International Supercomputing Conference (ISC’15), Frankfurt, Germany, July, 2015.
  4. Naoto Sasaki, Kento Sato, Toshio Endo and Satoshi Matsuoka, "Exploration of Application-level Lossy Compression for Fast Checkpoint/Restart", In HPC in Asia Workshop in conjunction with the International Supercomputing Conference (ISC'14), Leipzig, Germany, June, 2014.
  5. Kento Sato, Akira Nukada, Naoya Maruyama and Satoshi Matsuoka , "I/O acceleration with GPU for I/O-bound Applications", GPU Technology Conference 2014, Mar. 2014.
  6. Kento Sato, Adam Moody, Kathryn Mohror, Todd Gamblin, Bronis R. de Supinski, Naoya Maruyama and Satoshi Matsuoka, "Towards a Light-weight Non-blocking Checkpointing System", In HPC in Asia Workshop in conjunction with the International Supercomputing Conference (ISC'12), Hamburg, Germany, June, 2012.
  7. Kento Sato, Hitoshi Sato and Satoshi Matsuoka, "Model-based Optimization for Data-Intensive Application on Virtual Cluster", In the International Conference on Grid Computing (Grid 2008), Tsukuba, Sep, 2008.
  8. 佐藤賢斗, 佐藤仁, 松岡聡, "仮想クラスタを用いた Data-Intensive Application 実行環境の性能モデル構築と最適化に向けて", 先進的計算基盤システムシンポジウム 2008 (SACSIS2008), Tsukuba, June, 2008.

Posters (Non-refereed)

  1. Kento Sato,Adam Moody,Kathryn Mohror,Todd Gamblin,Bronis R. de Supinski, Naoya Maruyama and Satoshi Matsuoka, "Design and Modeling of a Non-Blocking Checkpoint System", In ATIP - A*CRC Workshop on Accelerator Technologies in High Performance Computing, Singapore, March, 2012.
  2. Kento Sato, Hitoshi Sato and Satoshi Matsuoka "Orchestrated Data Processing Acceleration for Data-Intensive Applications by using VM-based Migration", The 1st Data Intensive Science Workshop, Tokyo, March, 2011.

Talks

  1. Kento Sato, ``Extreme-Scale Resilience for Billion-Way of Parallelism'', ATIP Workshop: Japanese Research Toward Next-Generation Extreme Computing, Supercomputing 2014 (SC14), New Orleans, USA, November, 2014.
  2. Kento Sato, "APIs, Architecture and Modeling for Extreme Scale Resilience", Dagstuhl Seminar -14402, Dagstuhl, September 2014.
  3. 佐藤賢斗, "確率モデルとスーパーコンピュータ", 日本オペレーションズ・リサーチ学会 第248回待ち行列研究部会, Tokyo, July, 2014.
  4. Kento Sato, "Design and Modeling of an Asynchronous Checkpointing System", Emerging Technologies HPC Showcase, Supercomputing 2013 (SC13), Denver, USA, November, 2013.
  5. Kento Sato, "Checkpointing and Lustre", Japan LUG (Lustre Users Group) 2013, Tokyo, October, 2013.
  6. Kento Sato, "A Model-Based Algorithm for Optimizing I/O Intensive Applications in Clouds using VM-Based Migration", Korea-Japan E-Science Symposium, Miyagi, Aug, 2009.

Book Chapters

  1. Satoshi Matsuoka, Takayuki Aoki, Toshio Endo, Hitoshi Sato, Shin’ichiro Takizawa, Akihiko Nomura, and Kento Sato, "TSUBAME2.0: The First Petascale Supercom- puter in Japan and the Greatest Production in the World", volume 1, chapter 20, pages 525–556. Chapman & Hall/CRC Computational Science, April 2013. URL http://www.crcnetbase.com/doi/book/10.1201/b14677.