Page History

This section explains how the compute jobs you run on Gadi are charged.

On Gadi all compute jobs are charged based on the resources reserved for the job, and the amount of walltime used by the job. The resources reserved are currently calculated based on the maximum of either CPUs requested, or proportion of node memory requested.

This charging model allows jobs that would request significant memory in a node, with a small number of CPUs (effectively preventing other jobs from running on the remaining CPUs as there's no memory for other jobs) to be charged for the effective resources they are using. For the most cost effective job you should request only the resources you require for your job.

Job Cost (SU) = Queue Charge Rate  ✕  Max (NCPUs, Memory Proportion)  ✕  Walltime Used (Hours)

Queue Charge Rate = Charge rate for queue listed at Queue Limits. Note that express queues increase job priority, while also increasing the job's cost.

NCPUs = Number of CPUs requested for job with PBS -l ncpus request

Memory Proportion = Memory requested  ÷  Memory per core (Memory Per Node ÷ NCPUs per node for queue)