Workshop on Storage Challenges in the UK

The Special Interest Group of High-Performance I/O in the UK (SIG IO UK – Join our mailing list) organizes this workshop to bring together: users of data-intensive workloads, storage vendors, and system and middleware developers, in order to identify common challenges, to develop strategies for research and funding but also product development.

The workshop series generally covers all aspects of data access and management including I/O workflow handling, parallel file systems, middleware, tuning, performance monitoring, novel interfaces, storage technology, and data center perspectives. The workshop is primarily organized as a series of talks and group discussion slots.

This particular workshop focuses on exchanging information about challenges and the ongoing efforts to overcome them aiming to bring forward RD&E and the use of storage systems (in the UK). Users present their challenges faced dealing with I/O and ongoing RD&E to overcome them. Vendor talks focus on technical solutions for certain challenges accompanied by use cases demonstrating the benefit.

We welcome the submission of talks and short papers (in LaTeX article format) that summarize the challenges and ongoing effort in different institutions that we will lead to a joint paper submitted to Springer Lecture Notes in Computer Science as part of the ISC-HPC proceedings.

See our flyer.

This workshop is sponsored by DDN and powered by the Virtual Institute for I/O and ESiWACE ¹⁾.

The SIG-IO-UK is a loose consortium of people and institutions that share a common interest in High-Performance I/O. The SIG exchanges information revolving around the topic covering, but not limited to, I/O challenges, RD&E, and solutions on the market. We develop whitepapers raising awareness in the importance of I/O, describing the state of the art in the UK, aim to derive research and funding opportunities but also aid product development.

The workshop is organized by

• Julian Kunkel (Department of Computer Science, University of Reading, UK)

The following is the tentative agenda structure

• 09:30 Welcome coffee
• 09:45 Welcome, Status of the Next-Generation Interfaces – Julian Kunkel (Department of Computer Science, University of Reading)
  Slides
• 10:00 Challenges of benchmarking HPC storage systems – Jean-Christophe Rioual (MetOffice)
  Slides
• 10:30 In-Situ Post-Processing with the XIOS framework in the LFRic project – Samantha Adams (Met Office)
  Slides
• 11:00 DDN leverages years of HPC knowledge to fuse ingredients of AI thanks to A3I into traditional HPC workloads – Gautier Soubrane (DDN)
  Slides
• 11:30 WekaIO Matrix: Extracting the performance available from modern HPC hardware – Chris Weeden (Weka IO)
  Slides

• 12:00 Lunch, sponsored by DDN
  
• 13:00 Analysis of I/O usage on the ARCHER UK National Supercomputing Service – Andrew Turner (EPCC)
  Slides
• 13:30 Adventures in NoSQL for scale up computing – Jay Lofstead (Sandia National Laboratories)
  Slides
• 14:00 The Data Accelerator: Application and system performance for Burst Buffers on Cumulus – Alasdair J. King (Cambridge University)
  Slides

• 14:30 Coffee break
  
• 15:00 LASSI: Metric based IO analytics for HPC – Karthee Sivalingam (Cray)
  Slides (please write the author)
• 15:30 I/O research on the Astra ARM cluster – Matthew Curry (Sandia National Laboratories)
  Slides
• 16:00 Analysing I/O-Characteristics of Parallel Jobs – Eugen Betke (DKRZ)
  Slides
• 16:30 Cross-Site Data Movement – Jens Jensen (UKRI/STFC)
  Slides

• 17:00 Adjourn

Please use the University store to book (free attendance).

The official deadline for registration is Jan. 20th; but you may check if seats are available. Send an email to j.m.kunkel@reading.ac.uk; check that you get a confirmation.

Speakers must check the universities code of conduct and comply with these rules.

Jean-Thomas Acquaviva successively worked for Intel, the University of Versailles and the French Atomic Commission (CEA). He participated to the creation of their joint laboratory the Exascale Research Centre, where he led the Performance Evaluation Team. Today he’s actively contributing the development of the DDN Storage Advanced Technology Centre in France. Jean-Thomas has hands-on experience in the HPC ecosystem, with positions in start-up, large research public institutions, SME or global organizations. He has been a direct contributor to the ETP4PH Strategic Research Agenda and he’s chairing two conferences on high-performance storage.

Julian Kunkel is a Lecturer at the Computer Science Department at the University of Reading. He manages several research projects revolving around High-Performance Computing and particularly high-performance storage and is member of the Advanced Computing for Environmental Science (ACES) research group. Besides his main goal to provide efficient and performance-portable I/O, his HPC-related interests are: data reduction techniques, performance analysis of parallel applications and parallel I/O, management of cluster systems, cost-efficiency considerations, and software engineering of scientific software.

Jean-Christophe Rioual is a Scientific Systems Managers at the Met Office Hadley Centre for Climate Science and Services. He leads a team of scientific software engineers that develops and supports HPC applications and workflows. He is involved in the management, procurement and delivery of Met Office HPC systems.

Alasdair King is a Research Solutions Specialist at the University of Cambridge Research Computing Services. He currently works on the I/O performance and advanced filesystems work for the University, and a member of the team that supports Cumulus, a multi disciplinary service for STFS and EPSRC.

Karthee Sivalingam is a Research Engineer at the CRAY EMEA Research Lab. He is part of the Cray Center of Excellence for ARCHER that engages with users to allow them to maximise their use of Cray technologies. He has a particular interest in IO, Workflows, Optimisation, overlap of HPC with Big data and AI.

Samantha V. Adams is a Senior Scientific Software Engineer in the LFRic team within the UK Met Office. The LFRic project is developing the next generation of scalable weather and climate modelling infrastructure for HPC architectures of the future. Previously she worked as a Post-Doctoral Researcher in biologically inspired computing, encompassing Machine Learning and AI. Her current work involves parallel I/O in weather and climate systems and applying Machine Learning to weather and climate data.

Chris Weeden is a Senior Systems Engineer at WekaIO, the world’s fastest parallel file system company. Initially, he worked in Civil Engineering with many varied tasks from burning HGVs inside the arctic circle to filling the channel tunnel with smoke. He evolved into a systems administrator at WS Atkins taking on roles at Telewest and Sun Microsystems before becoming a Systems Engineer for Panasas in 2005 where he worked for 10 very happy and productive years installing many PBs into the field of HPC. After working a few years with Lustre and GPFS at Seagate and Cray, he was attracted to WekaIO by the highly innovative approach to high-performance storage.