r4 - 08 Feb 2019 - 10:45:15 - FredericDERUEYou are here: TWiki >  Atlas Web  >  Informatique > GuideAnalyse > GPUUserGuide

GPU user guide

This page is meant to describe the usage of HPC (High Performance Computing) and in particular GPU computing resources.

Use of GPU

How to use GPU at CC-IN2P3

At CC-IN2P3 you can access a GPGPU farm (under CentOS7), with 40 Nvidia Tesla K80 (10 machines of 4 GPU 16 cores) among which 2 machines are accessibles in interactive way.

To access these resources please open a ticket. By default you can access the interactive machines. For access to batch system, you need to check with your CAF representative at first, and in particular have an estimate of your needs (in number of hours).

How to use GPU at CERN

See documentation at the TechLab and the Forum on Concurrent Programming Models and Frameworks].

HPC resources in your laboratory

Several laboratories have access to similar resources directly in the lab or through the universities :

Other developments at IN2P3 and IRFU

Are given here links to existing joint efforts at IN2P3 and IRFU to develop the use of HPC. These are not specific developments to ATLAS:
  • Ecole informatique IN2P3 2016 : Parallélisme sur Matériel hétérogène
  • Atelier Activités et vision pour le domaine HTC / HPC: here
  • IN2P3 group for intensive coding: here
  • Master projet IN2P3 ComputeOps (APC, EC-Nantes, IPHC, LAL, LLR, LPNHE) on intensive computing using conteners (docker, rocket, lxd, udocker, singularity, shifter).

-- FredericDERUE - 01 Dec 2018

Edit | WYSIWYG | Attach | Printable | Raw View | Backlinks: Web, All Webs | History: r4 < r3 < r2 < r1 | More topic actions
Atlas.GPUUserGuide moved from Atlas.HPCUserGuide on 02 Dec 2018 - 16:55 by FredericDERUE - put it back
Powered by TWiki
This site is powered by the TWiki collaboration platformCopyright © by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding TWiki? Send feedback