User Guide

Policies, procedures and general documentation for the PACE Community Cluster as well as other HPC Clusters managed by the PACE group

The Job Scheduler: While test and development jobs can be run interactively on pacemaker, longer "production" job runs should be run in a batch mode across a larger segment of the cluster. Batches of jobs are managed by entities known as job schedulers and resource managers. Not only do these tools improve the use of the cluster across a broader workload but they also ensure that your job gets uninterrupted use of the processors, memory, and bandwidth scheduled for it during that run. Job scheduling for PACE manged clusters is done via the Moab scheduler from Cluster Resources the Torque resource manager from Altair. Those who have used OpenPBS or PBSPro will notice the similarities.
MPI: A brief tutorial on how to use MPI on PACE managed clusters
Cluster performance monitoring: Tools for inspecting performance of PACE managed clusters
Applications: A list and basic usage of applications installed on PACE managed clusters
PACE Infrastructure: A description of server and network infrastructure supporting PACE managed clusters
FAQ: Frequently asked Questions - and answers!
Cluster Descriptions: Text and graphics that describe and illustrate the various clusters operated by PACE
Links to external resources: Generally useful items, including documentation, developed outside of PACE.