HPC Services User Support Home

This is the Home page for the HPC User Support services offered by the High Performance Computing Services Group at Lawrence Berkeley National Lab and is intended for users of the Lab's Lawrencium cluster and users of PI-owned clusters managed by HPC Services. General information about HPC at Berkeley Lab can be found at our main web site here.

Our HPC User Support consultants have experience in computational science and engineering on a variety of parallel platforms and are available to assist users with issues related to using LBL computational resources. Users are encouraged to investigate our documentation to answer questions about our resources.

Please review the documentation links on the left side bar and click on the appropriate cluster link you need information about. You can reach our group for additional information by sending email to hpcshelp@lbl.gov Monday to Friday, 8:00am to 6:00 pm.


Jul 28, 2016 - No Runtime Limits on Free Low Priority QoS
After evaluating user feedback, we have decided to remove the 64-node and 72-hour time limit for the jobs that one can run through the lr_lowprio and mako_lowprio QoSs to better serve our user community. This means there will be essentially no limit with the resource you can request at a time; however, please keep it in mind that preemption still applies whenever resources are demanded by other higher priority jobs.

Mar 22, 2016 - Run on Lawrencium for Free - New Low Priority QoS
We are pleased to announce the “Low Priority QoS (Quality of Service)” pilot program which allows all users to run jobs requesting up to 64 nodes and up to 3 days of runtime on the Lawrencium Cluster resources at no charge when running at a lower priority. Users should check the Lawrencium user page for specific instructions for submitting jobs to the new QoS.

Nov 16, 2015
We are pleased to announce the availability of the Lawrencium LR4 partition consisting of 96 ea. Haswell processor compute nodes. The new nodes, which feature two Intel Haswell E5-2670v3 12-core processors running at 2.3 Ghz, are capable of executing twice as many floating point instructions per clock cycle as compared to the previous LR3 generation IvyBridge nodes.