BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Drupal//recurring_events_ical//2.0//EN
BEGIN:VEVENT
UID:49607bee-ad70-42ab-9599-4e7790df4ba8@support.access-ci.org
DTSTAMP:20250102T134755Z
DTSTART:20250408T180000Z
DTEND:20250408T193000Z
SUMMARY:Porting your AI application to Voyager Gaudi Architecture (Advanced
  HPC-CI Webinar Series)
DESCRIPTION:This tutorial will cover how to run a Deep Learning (DL) model,
  based on the Pytorch framework, on the Voyager supercomputer. Voyager com
 prises an innovative architecture to optimize DL applications thanks to it
 s Intel Habana Gaudi accelerators. We will discuss Voyager architecture an
 d show how to launch jobs using containers in Kubernetes. Then, we will ta
 ke a Pytorch application and will port it using Intel Gaudi libraries. We 
 will discuss how to run Jupyter notebooks on Voyager. Finally, we will sho
 w how to pull a Hugginface model and run inference on Voyager using Transf
 ormers/Diffusers and Optimum-Habana libraries.InstructorDr. Javier Hernand
 ez-Nicolau is a Data Scientist in the Scientific applications group at SDS
 C. He obtained his PhD in Plasma Physics and Nuclear Fusion at the Univers
 ity of Carlos III of Madrid (Spain) in 2019 and then worked as a Project S
 cientist at the University of California-Irvine. He developed and optimize
 d MHD and gyrokinetic fusion plasma codes for several HPC systems using Fo
 rtan, OpenMP and OpenACC. He joined SDSC in 2023 where he keeps collaborat
 ing with the Nuclear Fusion community developing ML/AI applications. He al
 so offers user support to deploy a large variety of AI models on the Voyag
 er machine which uses Intel Gaudi accelerators.See a full list of SDSC's u
 pcoming training and events here.
URL:https://support.access-ci.org/events/7723
END:VEVENT
END:VCALENDAR