Computational Systems
NERSC provides large-scale, state-of-the-art computing systems for its users.
Hopper
Hopper, a Cray XE6, with 153,216 compute cores, 217 TB of memory and 2PB of disk is the largest system at NERSC. Read More »
Edison
NERSC's new system, code-named "Cascade" by Cray and "NERSC 7" at NERSC, will have a peak performance of greater than two petaflops per second (2 PF) when installed in 2013. The integrated storage system will have more than six petabytes (6 PB) of storage with an I/O bandwidth of 140 GB per second. Read More »
Carver
Carver, an IBM iDataPlex system, provides excellent CPU performance, good energy efficiency per flop, and a familiar environment for mid-range parallel applications. Carver's computational nodes contain 2240 Intel Nehalem quad-core processors, 160 Westmere 6-core processors, and 8 Nehalem-EX 8-core processors, for a total of 9,984 cores. Read More »
Euclid - Retired 01/31/2013
Euclid was retired from NERSC service on 31 January, 2013. View this web page to see migration options. Read More »
Dirac GPU Cluster
Dirac is a GPU testbed cluster run in collaboration with Berkeley Lab's Computational Research Division. NERSC users may apply for access to the Dirac system. Read More »
PDSF
PDSF is a networked distributed computing cluster designed primarily to meet the detector simulation and data analysis requirements of physics, astrophysics and nuclear science collaborations. Read More »
Genepool
The Genepool system is a cluster dedicated to the DOE Joint Genome Institue's computing needs. Read More »
