Skip to end of metadata
Go to start of metadata
You are viewing an old version of this page. View the current version. Compare with Current  |   View Page History

Common GPU Productivity Environment

GPU productivity environments have recently become an important part of scientific computing and visualization. The following is an attempt to define which ones are relevant for the LinkSCEEM-2 community and set an action plan.

BC Policy: LS2_11-99
Date of Policy: 1st January 2012
First Update: 1st December 2011
Second Update: 2nd December 2011

The main objective of this policy is to provide the following common GPU productivity environment across all LinkSCEEM-2 resources:

Package/Tool Requested version(s) Compliance level
CUDA Toolkit v4.0 MUST
pyCUDA 2011.2.1 MUST
pyOpenCL 2011.2 MUST
MAGMA v1.1 MUST
ViennaCL 1.2.0 MUST
AMD math library for OpenCL (APPML) v1.4 MUST
Intel OpenCL SDK v1.5 SHOULD
CULA r12 SHOULD
gputools R package v0.26 SHOULD
rCUDA v3.1 SHOULD

These GPU productivity environment will be supplemented with other such productivity tools as they become available on allocated systems.

Labels:
None
Enter labels to add to this page:
Please wait 
Looking for a label? Just start typing.