Versions Compared

Old Version 25

changes.mady.by.user Malte Thoma

Saved on

compared with

New Version 26

changes.mady.by.user Malte Thoma

Saved on

Key

  • This line was added.
  • This line was removed.
  • Formatting was changed.

...

Partition

Nodes

Description

smp

prod-[001-240]

  • default partition,

  • MaxNodes=1 → MaxCores=128,

  • default RAM: 1900 MB/core

  • Jobs can share a node

mpp

prod-[001-240]

  • exclusive access to nodes,

  • MaxNodes=240

fat

fat-00[1-4]

  • like smp but for jobs with extensive need of RAM

  • default RAM: 30000 MB

gpu

gpu-00[1-2]

  • like smp but...

  • ... you have to specify the type and number of desired GPUs via --gpus=<GpuType>:<GpuNumber> .
    The two gpu nodes each contain a different number and type of GPU:

    • gpu-001: 2x a40

    • gpu-002: 4x a100

Quality of service (

...

--qos)

A higher priority means your job is scheduled before other jobs. In addition,  during working hours 10 nodes are reserved exclusively for jobs using qos=30min (to facilitate development and testing). For longer runs, another QOS (and walltime) has to be specified.  Note: long running jobs (longer than 12 hours, up to 48 hours) “cost” more in terms of fairshare (meaning you priority will decrease for further jobs).

...