At the UCSF Helen Diller Family Comprehensive Cancer Center we have a large Linux compute cluster that is available to anyone affiliated with the Cancer Center - feel free to contact us if you would like to join or have questions.
As of February 2017, the cluster has 28 nodes and ~1,100 cores, which predominantly are AMD processors. Each node has up to 512 GiB of RAM and up to 7 TiB of fast local disk space. In addition to ~90 TiB global disk space shared among all users, several research groups have disk space of their own mounted to the cluster. The cluster uses Scyld ClusterWare, which for instance means that all nodes have identical Linux setups (Red Hat Enterprise 6.6) and identical software tools installed. We use TORQUE PBS for scheduling jobs and Moab to manage the workload.
For further details on the cluster and how to use it, see https://ucsf-ti.github.io/tipcc-web/.