BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Drupal//recurring_events_ical//2.0//EN
BEGIN:VEVENT
UID:a5d810f0-94a6-4baf-872c-e36c28e20e97@support.access-ci.org
DTSTAMP:20240308T081647Z
DTSTART:20240312T180000Z
DTEND:20240312T200000Z
SUMMARY:Efficient I/O: HDF5 and DLIO Workshop
DESCRIPTION:This workshop will include two talks about optimizing, profilin
 g, and benchmarking input/output operations using HDF5 and DLIO.Optimizin
 g Your I/O Workload: Techniques for Effective HDF5 Usage - Scot Breitenf
 eld, The HDF GroupHDF5 is a widely used data model, file format, and I/O l
 ibrary, particularly in HPC applications for managing and storing large am
 ounts of simulation data. This talk will focus on HDF5 usage on NCSA's Del
 ta system and provide a brief overview of HDF5 (serial and parallel) with 
 an emphasis on HDF5 HPC tuning techniques such as collective metadata I/O,
  data aggregation, asynchronous I/O, and other HDF5 tuning parameters and 
 features. During the presentation, we will discuss various storage options
  available on Delta for HDF5 files and multiple post-simulation storage op
 tions, including cloud storage and other tools.Deep Learning I/O: Benchmar
 k and Profiling - Hariharan Devarajan, Lawrence Livermore National Laborat
 oryIn this talk, we will be taking a dive into the unique features of I/O 
 in deep learning application. We will present our ongoing efforts to under
 stand the I/O characteristics of DL workloads using our DLIO profiler. Fin
 ally, we will present the DLIO Benchmark which is built to accurately repr
 esent the I/O characteristics in deep learning workloads.Date and location
 : these talks will be presented virtually via Zoom on 3/12/2024, from 1 t
 o 3 PM CT.
URL:https://support.access-ci.org/events/7452
END:VEVENT
END:VCALENDAR