HPC Wiki - The U of A HPC Wiki is the user's guide to high-performance computing at the University in a familiar and easy to access format.

  Listserv - The U of A HPC Wiki is the best place to discuss problems or questions. Simply email questions to the listserv or become a member and find out what others are talking about?

Learn the HPC Lingo

Get to know your supercomputing jargon with our handy list of frequently used words, phrases and acronyms on the hpcWiki.

How to Use the Clusters

AHPCC computing resources consist of two clusters: Trestles, an 8,192 core Appro cluster with a peak performance of 79 TF (Trillion Floating Point Operations per Second), and Razor, a 4,328 core IBM/Dell cluster with a peak performance of 76 TF. The clusters are augmented with special purpose nodes with another 624 cores and 14 TF: seven NVidia GPU nodes and ten large memory nodes consisting of one 3072 GB Intel, three 768 GB Intel, and six 512 GB AMD nodes. Water cooling doors are installed on racks which have the highest head load.

Supported Software

Each cluster has a variety of software installed in support of users. A complete listing of software by cluster is detailed on on the Software page, and is updated each hour from the clusters to capture any newly installed packages. Instructions for using supported software are added to the HPCWiki as time permits.