IPDPS'20

Welcome


The IEEE International
Workshop on
High-Performance Storage

In conjunction with the 34th IEEE International Parallel and distributed processing symposium (IPDPS 2020)

Covid-19

Due to the current situation, the IPDPS conference and the attached workshops will not take place as planned.
  • The conference proceedings, including the workshop proceedings, will be published as originally planned.
  • Presentation materials (slides) will be posted on this web site
  • Other details will be published as they become available




Welcome to HPS 2020

The recent years are seeing an accelerated evolution of high-end storage systems, libraries, and services, due to several reasons: 

Technology: The availability of an increasing number of persistent solid-state storage technologies that can replace either memory or disk is creating new opportunities for the structure of storage systems.

Performance Requirements: Disk-based parallel file systems cannot satisfy the performance needs of high-end systems. It is not clear how solid-state storage is best used to alleviate the problem.

Infrastructure evolution: HPC technology will not only be deployed in dedicated supercomputing centers in the future. “Embedded HPC”, “HPC in the box”, “HPC in the loop”, “HPC in the cloud”, “HPC as a service”, “near- to-real-time simulation” are concepts requiring new small-scale deployment environments for HPC. A federation of systems and functions with consistent mechanisms for managing I/O, storage and data processing across all participating systems will be required, to creating a “continuum” of computing.

Application Evolution: Data analysis, including graph analytics and machine learning training, is becoming an increasingly important high-end applications  I/O is often a major bottleneck for such an application, both in a cloud environment and in an HPC environment – especially when fast turnaround or integration of heavy computation with heavy analysis are required.

Virtualization and disaggregation: As virtualization and disaggregation become broadly used in cloud and HPC computing, the issues of virtualized storage and storage disaggregation have an increasing importance.

The workshop aims to bring together researchers and developers working on high-end storage systems, libraries and applications in HPC and clusters, and users of such systems that are interested in these issues.

Call for Papers

HPS 20120 welcomes original submissions in a range of areas, including but not limited to:

  • High-end storage systems
  • Parallel and distributed high-end storage organizations
  • The synergy between different storage models (POSIX file system, object storage, key-value, row-oriented and column-oriented databases)
  • HPC/Cloud/Edge infrastructures
  • Structures and interfaces for leveraging persistent solid-state storage
  • High-performing I/O libraries and services
  • I/O performance in high-end systems and applications
  • Benchmarks and performance tools for high-end I/O
  • Language and library support for data-centric computing
  • Storage virtualization and disaggregation
  • Active processing in storage technologies.
  • Ephemeral storage media and consistency optimizations
  • Storage architectures and systems for scalable stream-based processing
  • Study cases of I/O services in support of various application domains (bioinformatics, scientific simulations, large observatories, experimental facilities, etc.)

Papers should present original research and provide sufficient background material to make them accessible to the broader community.

Draft Agenda

9am-10am
Intro & Keynote

10am    -    10:30am
Coffee Break
10:30am -   12:30am
Paper Session 1: I/O optimization

Optimizing Asynchronous Multi-level Checkpoint/Restart Configurations with Machine Learning. T.  Dey, K. Sato, B. Nicolae, J. Guo, J. Domke, W. Yu, F. Cappello and K. Mohror

On Overlapping Communication and File I/O in Collective Write Operation. R. Feki and E. Gabriel

Enhancing Endurance of SSD Based High Performance Storage Systems using Emerging NVM Technologies. T. Roy and K. Kant

Design of Locality-aware MPI-IO for Scalable Shared File Write Performance. K. Sugihara and O. Tatebe
12:30am   -      2pm
Lunch
2pm         -      3pm
Panel
3pm          -  3:30pm
Coffee Break
3:30pm     -     5pm
Paper Session 2 - Understanding and managing storage

Dynamic Provisioning of Storage Resources: A Case Study with Burst Buffers. F. Tessier, M. Martinasso, M. Chesi, M. Klein and M. Gila

Recorder 2.0: Efficient Parallel I/O Tracing and Analysis. C. Wang, J. Sun, M. Snir, K. Mohror and E. Gonsiorowski

Silent Data Access Protocol for NVRAM+RDMA Distributed Storage. Q, Liu and P. Varman

Keynote Talk

Speaker: Jay Lofstead
Abstract: TBD
Short Bio TBD

Panel

Topic: New storage technologies – how should they be used in high-end I/O systems?
Panel members - TBD

Submission Info

All submissions should follow the IEEE standard 8.5” x 11” two-column format. The workshop will accept traditional research papers (8-10 pages) for in-depth topics and short papers (4-8 pages) for works in progress on hot topics.

  • Long papers: 8-10 pages, with a full problem description, background and related work, design, and evaluation.
  • Short papers: 4-8 pages, for works in progress on hot topics.

All the papers should be submitted here.

Papers should not be submitted in parallel to any other conference or journal.

The proceedings of this workshop will be published together with the proceedings of other IPDPS 2020 workshops by the IEEE Computer Society Press. Proceedings of the workshops are distributed at the conference and are submitted for inclusion in the IEEE Xplore Digital Library after the conference. At least one of the authors of each accepted paper must register as a participant of the workshop and present the paper at the workshop, in order to have the paper published in the proceedings.

Camera-Ready Submission

Please follow emailed instructions and check the main conference website.

Important Dates

  • Abstract submission (optional) deadline: January 15th, 2020
  • Paper submission deadline: January 31st, 2020   Extended to Friday Feb 7th.
  • Acceptance notification: February 28th, 2020
  •  Camera-Ready deadline: March 15th, 2020
  •  Workshop: May 22th, 2019

Organizing Committee

  • Gabriel Antoniu , INRIA Renne     
  • Franck Cappello, Argonne National Laboratory
  • Tony Cortes, Barcelona Supercomputing Center
  • Kathryn Mohror, Lawrence Livermore National Laboratory
  • Kento Sato, RIKEN
  • Marc Snir, University of Illinois at Urbana-Champaign
  •  Weikuan Yu, Florida State University

Workshop Co-chairs

  • Kathryn Mohror
  • Marc Snir (chair@hpsworkshop.org)

Program Co-chairs

  • Gabriel Antoniu
  • Tony Cortes

Program Committee

  • John Bent, Seagate Systems, USA
  • Angelos Bilas, Forth, Greece
  • André Brinkmann, U Mainz, Germany
  • Suren Byna, LLBL, USA
  • Franck Cappello, ANL, USA
  • Jesus Carretero, Universidad Carlos III de Madrid, Spain
  • Alexandru Costan, Inria and INSA Rennes, France
  • Matthieu Dorier, Argonne National Lab, USA
  • Carlos Maltzan, University of California, Santa Cruz
  • Pierre Matri, Universidad Politécnica de Madrid, Spain
  • Bogdan Nicolae, Argonne National Lab, USA
  • Manish Parashar, Rutgers University, USA
  • Maria Pérez, UPM, Spain 
  • Dana Petcu, University West Timisoara, Romania
  • Rob Ross, Argonne National Laboratory, USA
  • Michael Schoettner, University of Dusseldorf, Germany
  • Xuanhua Shi, Huazhong University of Science and Technology, China
  • Domenico Talia, University of Calabria, Italy
  • Osamu Tatebe, University of Tsukuba, Japan
  • Kento Sato, RIKEN, Japan
  • Weikuan Yu, Florida State University, USA

Registration and Hotel

The workshop does not have a separate registration. All attendees need to register with the main conference. Details about registration and hotel can be found on the main conference website.

Visa

 

Please check the main conference website.