dib-lab / farm-notes

notes on the farm cluster
16 stars 9 forks source link

Farm resources we have access to #41

Open SichongP opened 2 years ago

SichongP commented 2 years ago

Under allocation limits, it says:

As of June 15, 2020, they are as follows.

On bmh/bmm/bml, we have one bigmem node, so:

  • 1TB of memory max
  • 96 CPU

On high2/med2/low2, we have two parallel nodes, so:

  • 512 GB RAM total, 256 GB max per job
  • 32 cores on one machine, 64 cores on the other machine

I've been having a hard time queueing on bmm lately so I asked admin and here is what they said:

You also have access to the bmh partition. In that partition, ctbrowngrp has 1 minute access to 48 CPUs and 500 GB of RAM.

Through ctbrowngrp you also have access to med2 and high2 with 224 CPUs and 500 GB of RAM.

Is this inaccurate or has there been changes in our buy-ins?