Several queues (or “partitions”) are availble to users for different types of jobs. One queue is restricted to single-node (serial or threaded) jobs; another for multinode parallel programs, and others are for access to specialty hardware such as large-memory nodes or nodes offering GPUs.

Partition Max time / job Max nodes / job Max cores / job Max cores / node Max memory / core Max memory / node / job SU Charge Rate
standard 7 days 1 40 40 9GB 375GB 1.00
parallel 3 days 25 1000 40 9GB 375GB 1.00
largemem 4 days 1 16 16 64GB 975GB 1.00
gpu 3 days 4 10 10 32GB 375GB 3.00 *
knl 3 days 8 512 cores / 2048 threads 512 3GB (per physical core) 192GB 1.00
dev 1 hour 2 8 4 6GB 36GB 0.00

* GPU charge rate = number of cores + 2 * number of GPU devices.