Agenda

The workshop will be held in Heidelberg, Germany on Tuesday, September 6, 2022 in room HS 6.

Note: All presenters and attendees must register at the IEEE Cluster website.

Agenda: (all times are in Central European Summer Time / CEST)

09:15 - 9:30 – Welcome Message & Speed Introduction
09:30 - 10:30 – Keynote: Organic HPC - Processing on Heterogeneous (Storage) Landscapes [Slides]
- - - Speaker: Julian Kunkel, Georg-August-Universität Göttingen/GWDG, Germany
10:30 - 11:00 – Coffee Break
11:00 - 12:30 – Session I: Research Papers
- - - Paper 1: Protecting Metadata Servers From Harm Through Application-level I/O Control, Ricardo Macedo, Mariana Miranda, Yusuke Tanimura, Jason Haga, Amit Ruhela, Stephen Harrell, Richard Todd Evans and João Paulo. INESC TEC & University of Minho, Portugal, National Institute of Advanced Industrial Science and Technology (AIST), Japan, Texas Advanced Computing Center (TACC), USA and Intel, USA. [Slides]
    - Paper 2: A Comprehensive I/O Knowledge Cycle for Modular and Automated HPC Workload Analysis, Zhaobin Zhu, Sarah Neuwirth and Thomas Lippert. Goethe University Frankfurt, Germany and Jülich Supercomputing Centre, Germany. [Slides]
    - Paper 3: Assessment of the I/O and Storage Subsystem in Modular Supercomputing Architectures, Sarah Neuwirth. Goethe University Frankfurt, Germany and Jülich Supercomputing Centre, Germany. [Slides]
12:30 - 14:00 – Lunch Break
14:00 - 15:30 – Session II: Expert Talks
- - - Expert Talk I: PMEM and NVMe for storage: How do the Different Storage System Approaches Affect Performance?, Jay Lofstead, Sandia National Laboratories, USA. [Slides]
    - Expert Talk II: Towards Building a User Centric I/O Performance Tuning Workflow, Radita Liem, RWTH Aachen, Germany. [Slides]
    - Expert Talk III: Scalable I/O for Node-local Storage in HPC for DL and Checkpointing Workloads, Awais Khan, Oak Ridge National Laboratory, USA. [Slides]
15:30 - 16:00 – Coffee Break
16:00 - 17:00 – Session III: Expert Talks Part 2
- - - Expert Talk IV: Composing for Supercomputers: HPC, Data Services, and the Mochi Project, Matthieu Dorier, Argonne National Laboratory, USA. [Slides]
    - Expert Talk V: Challenges in Characterizing I/O in Large Scientific Workflows, Hariharan Devarajan, Lawrence Livermore National Laboratory, USA. [Slides]
17:00 - 17:30 – Discussion: Emerging HPC Workloads and the Future Direction of Parallel I/O Research