Lustre User Group 2021

From Lustre Wiki
Revision as of 17:36, 20 May 2021 by Adilger (talk | contribs) (add events link)
Jump to navigation Jump to search

LUG 2020 was to be held in Gainesville, FL on May 19th-20th at the University of Florida Research Computing.

The event was held virtually due to the ongoing COVID-19 pandemic.

LUG 2021 Agenda

Wednesday, May 19th

9:00-9:05am Opening Remarks
OpenSFS, University of Florida
9:05-9:35am Lustre Community Release Update (video TBD)
Peter Jones, Whamcloud
9:35-10:05am Status of the Linux Native Lustre Client (video TBD)
James Simmons, ORNL
10:05-10:35am A Triad based Architecture for a Multipurpose Lustre Filesystem at /rdlab (video TBD)
Gabriel Verdejo Alvarez, Univeritat Politecnica de Catalunya (UPC)
10:35-11:05am Impact of ZFS Direct I/O on OSD Performance (video TBD)
Christopher Brumgard, ORNL
11:15am Efficient Metadata Scanning using ZFS and NVMe-over-Fabrics (video TBD)
Ross Miller, ORNL
11:45am Break
11:45am Optimizations for AI in the EXAScaler Filesystem (video TBD)
James Coomer, DDN
12:00pm File Level Erasure Coding in Lustre (video TBD)
Adam Disney, ORNL
12:30pm Lustre at Scale: ALCF Update (video TBD)
Alex Kulyavtsev, Gordon McPheeters, ANL
1:00pm Virtual Networking Session

Thursday, May 20th

9:00am Opening Remarks
OpenSFS, University of Florida
9:05am Accelerating AI at-scale with Selene DGXA100 SuperPOD and Lustre Parallel Filesystem Storage (video TBD)
Prethvi Kashinkunti and Julie Bernauer, NVIDIA
9:35am Lustre 2.15 and Beyond (video TBD)
Andreas Dilger, Whamcloud
10:05am Lustre Client Encryption (video TBD)
Sébastien Buisson, Whamcloud
10:35am Update on “Secure Lustre” (video TBD)
James Beal, Pavlos Antoniou, Wellcome Sanger Institute UK
11:05am Break
11:30am Evaluation of DoM/DNE scaling performance in Lustre (video TBD)
James Simmons, ORNL
12:00pm Giving Easy User Access to Lustre JobStats and Robinhood Information (video TBD)
Simon Guilbault, Université Laval
12:30pm OpenSFS Update and Annual Meeting
OpenSFS
1:00pm Virtual Networking Session