This is an old revision of the document!
All jobs on AHPCC clusters which require a significant amount of CPU or memory should be submitted through the queueing system. In general, two types of jobs may be passed into the queue:
A batch job - a specific command is executed on the node(s) assigned to the job without the need for user interaction. A vast majority of jobs ran on the HPC clusters are batch jobs.
An interactive job - a login shell is started on the first node assigned to the job. The user, in turn, specifies the commands to execute at the command prompt.
All compute nodes in the Razor and Trestles clusters are divided into groups of nodes called queues. All nodes assigned to a particular queue are identical. The queues differ from each other by the following factors:
type of cpu and number of cores on each node
number of nodes assigned
the maximum number of nodes allowed to be used by a single job
amount of memory
walltime - the maximum amount of execution time for a single job
Queues - summary of public queues