The fees for HPC resource usage are defined as follows:
1) ZEUS Cluster
Accounting on the ZEUS cluster is based on PBS reports. Charges apply only to public queues (those with names starting with “zeus_”).
CPU usage – 0.015 NIS per core-thread per hour
GPU usage – 0.18 NIS per GPU card per hour
At the beginning of each month, all HPC users receive detailed reports of their PBS jobs that ran during the previous month.
A budget owner (PI) receives reports containing the CPU/GPU usage for each member of their group, as well as the total fees for the previous month.
They also receive a message with payment instructions.
Every user can check their current PBS usage for the month by using the `prj_usage` command.
2) ATHENA and DGX Clusters
Accounting on the ATHENA and DGX clusters is based on SLURM reports. Charges apply only to public/shared resources.
GPU usage – 0.18 NIS per GPU card per hour
CPU usage – 0.015 NIS per core-thread per hour
CPU usage on GPU clusters is charged if a job runs on CPU only.
At the beginning of each month, all HPC users receive detailed reports of their SLURM jobs that ran during the previous month.
A budget owner (PI) receives a report containing the CPU/GPU usage for each member of the project group and the total fees for the previous month.
They also receive a message with payment instructions.