Code Together

Maximum Performance Storage for AI/HPC - How DAOS Enables Large Dataset Workloads on Aurora


Listen Later

Tony is joined by Kevin Harms of Argonne Leadership Computing Facility (ALCF) and Mohamad Chaarwi of Intel to discuss how the open source Distributed Asynchronous Object Storage (DAOS) project will enable world class I/O performance on the Aurora Exascale Supercomputer.

Guests:

Kevin Harms is a team lead of the ALCF's Performance Engineering group. His interests include working on high performance storage and I/O.

Mohamad Chaarawi is a senior software engineer in the DAOS group at Intel. He is responsible for the DAOS client library and other middleware IO libraries and application porting over the DAOS stack.

Kevin Harms

Mohamad Chaarawi

Learn More:
Aurora Supercomputer
https://www.alcf.anl.gov/aurora
 
Argonne Leadership Computing Facility
https://www.alcf.anl.gov
 
 Distributed Asynchronous Object Storage (DAOS)
https://docs.daos.io/v2.2/
 
IO500 List
https://io500.org/list/sc22/full?sort=io500_bw&direction=desc
 
oneAPI
https://oneapi.io
https://www.intel.com/content/www/us/en/developer/tools/oneapi/overview.html
 

Check out more Intel podcasts at: https://www.intel.com/podcasts

...more
View all episodesView all episodes
Download on the App Store

Code TogetherBy Code Together

  • 5
  • 5
  • 5
  • 5
  • 5

5

3 ratings