NERSCPowering Scientific Discovery Since 1974

Dirac GPU Cluster

Current Status: Up

NOTICE: Dirac will be shut down permanently and retired from service at 17:00 on Friday, December 12, 2014.

Dirac is a testbed experimental system and is not considered a NERSC production system. Therefore, policies regarding user access and staff support differ substantially from other NERSC systems. As Dirac is a sub-cluster of Carver, whenever Carver is down for maintenance Dirac will be down. Every effort will be made to give users advance notice in these situations.  Users who desire access to this resource must have research objectives that align with Dirac's research objectives.

Dirac has well served its intended purpose of allowing users to evaluate GPU technologies.  It will be removed from service on December 12, 2014.



Request for Dirac Account

Dirac is a 48-node GPU cluster intended to help researchers better understand the use of GPUs for scientific applications. Access is available to those who want to investigate the following: GPU computing using CUDA and/or OpenCL. Scaling codes on a multi-GPU system. Evaluate programming models for using GPUs in HPC, e.g. MPI + CUDA, OpenMP + CUDA, etc. To request access to Dirac, please fill out the form below. You can expect a response from NERSC within two weeks. Please contact the NERSC… Read More »

Node and GPU Configuration

Dirac is a 50 node GPU cluster with NVIDIA Fermi chips. Read More »

Access, Compiling and Running Jobs

Information on compiling CUDA codes and hybrid MPI+CUDA applications. Read More »