Difference between revisions of "Colibri cluster"
Jump to navigation
Jump to search
(cp in http://ccm.ucdenver.edu/wiki/Colibri_cluster) |
|||
| Line 7: | Line 7: | ||
* [[Colibri user's documentation]] | * [[Colibri user's documentation]] | ||
* [[Colibri administrator's documentation]] | * [[Colibri administrator's documentation]] | ||
| + | |||
| + | [[Category:Hardware]] | ||
Latest revision as of 16:01, 16 October 2021
This cluster was funded by NSF grant 0958354 GPU Cluster for Computing Research. , Principal Investigator Julien Langou, Co-PIs Jack Dongarra, Jan Mandel, Daniel Connors, and Jonathan Beezley.
The cluster consists of 24 compute nodes each equipped with two CPUs with 16 cores each. and two NVIDIA Fermi GPUs, connected by Infiniband. Please see the legacy cluster page for more information on the hardware. It was delivered by HP in 2013. In 2014, the license on software run out and the cluster was reinstalled with Rocks+ by StackIQ, which is the same cluster distribution as on the Gross cluster.