Shumpei Shiina

I am a researcher studying system software for parallel and distributed processing for large-scale data and computation. I obtained a Ph.D. degree at Taura Laboratory, Department of Information and Communication Engineering, Graduate School of Information Science and Technology, The University of Tokyo, Japan.

Research

My primary research interest is in parallel and distributed computing systems that enable easy and efficient computation on massively parallel computers, from multi-core CPUs to clusters or supercomputers.

Itoyori is a C++ library developed to that end. Itoyori enables highly productive global-view task-parallel programming, while still delivering high performance. If interested, please check out Itoyori’s documentation and GitHub.

Keywords: task parallelism, threads, dynamic load balancing, work stealing, PGAS, etc.

Publications

Conference Papers and Journal Articles (Peer Reviewed)

• Shumpei Shiina, Kenjiro Taura.
  Itoyori: Reconciling Global Address Space and Global Fork-Join Task Parallelism.
  The International Conference for High Performance Computing, Networking, Storage, and Analysis (SC ‘23). 2023.
  💡 Acceptance rate: 24% (90/376)
  [🔗 doi] [📄 Paper] [📑 Slides] [🔨 Artifacts]
• Shumpei Shiina, Kenjiro Taura.
  Improving Cache Utilization of Nested Parallel Programs by Almost Deterministic Work Stealing.
  IEEE Transactions on Parallel and Distributed Systems (TPDS). 2022.
  [🔗 doi] [📄 Paper]
• Shumpei Shiina, Kenjiro Taura.
  Distributed Continuation Stealing is More Scalable than You Might Think.
  The 24th IEEE International Conference on Cluster Computing (Cluster ‘22). 2022.
  💡 Acceptance rate: 24% (43/181)
  [🔗 doi] [📄 Paper] [📑 Slides] [🔨 Artifacts]
• Shumpei Shiina, Shintaro Iwasaki, Kenjiro Taura, Pavan Balaji.
  Lightweight Preemptive User-Level Threads.
  The 26th ACM SIGPLAN Symposium on Principles and Practice of Parallel Programming (PPoPP ‘21). 2021.
  💡 Acceptance rate: 21% (31/150)
  [🔗 doi] [📄 Paper] [📑 Slides] [🔨 Artifacts]
• Shumpei Shiina, Kenjiro Taura.
  Almost Deterministic Work Stealing.
  The International Conference for High Performance Computing, Networking, Storage, and Analysis (SC ‘19). 2019.
  💡 Acceptance rate: 25% (87/339)
  [🔗 doi] [📄 Paper] [📑 Slides]

Dissertations

• [Doctoral Dissertation] Task Parallelism at Scale: Locality-Aware Scheduling and Global Address Space.
  Graduate School of Information Science and Technology, The University of Tokyo. Mar. 2024.
  💡 Dean’s Award
  [📄 Paper]
• [Master’s Thesis] Efficient Thread Scheduling Strategies for Nested Parallel Programs.
  Graduate School of Information Science and Technology, The University of Tokyo. Mar. 2021.
  💡 Dean’s Award
  [🔗 doi]
• [Bachelor Thesis] Almost Deterministic Scheduling for Dynamic Task Parallelism.
  Department of Information and Communication Engineering, The University of Tokyo. Mar. 2019.
  💡 Outstanding Graduation Thesis Award

Awards

• 2024: Dean’s Award for Academic Achievement, Graduate School of Information Science and Technology, The University of Tokyo.
• 2023: IPSJ Yamashita SIG Research Award.
• 2021: Dean’s Award for Academic Achievement, Graduate School of Information Science and Technology, The University of Tokyo.
• 2019: Best Research Award, The 3rd cross-disciplinary Workshop on Computing Systems, Infrastructures, and Programming (xSIG) 2019.
• 2019: Outstanding Graduation Thesis Award, Department of Information and Communication Engineering, The University of Tokyo.

Biography

Education

• Apr. 2021 – Mar. 2024. Ph.D.
  • Department of Information and Communication Engineering, Graduate School of Information Science and Technology, The University of Tokyo.
  • Supervisor: Prof. Kenjiro Taura
• Apr. 2019 – Mar. 2021. Master of Information Science and Technology.
  • Department of Information and Communication Engineering, Graduate School of Information Science and Technology, The University of Tokyo.
  • Supervisor: Prof. Kenjiro Taura
• Apr. 2015 – Mar. 2019. Bachelor of Engineering.
  • Department of Information and Communication Engineering, The University of Tokyo.
  • Supervisor: Prof. Kenjiro Taura

Carrier

• Apr. 2024 – Present. Researcher at Toyota Motor Corporation.
• Apr. 2021 – Mar. 2024. JSPS Research Fellowship for Young Scientists (DC1).
  • Project: Dynamic Task Scheduling Strategies for Deep Memory Hierarchies in the Future
  • Grants: 2,200,000 YEN
• Sep. 2019 – Feb. 2020. Research Aide, Argonne National Laboratory.
  • Supervisor: Pavan Balaji, Mentor: Shintaro Iwasaki

Academic Activities

• The Thirteenth International Workshop on Accelerators and Hybrid Emerging Systems (AsHES ‘23), Web Chair.
• The 27th ACM SIGPLAN Symposium on Principles and Practice of Parallel Programming (PPoPP ‘22), Artifact Evaluation (AE) Committee.
• The Seventh Workshop on Accelerator Programming Using Directives (WACCPD ‘20), Web Chair.