37
pyaf
8d

So, my college has a server with a single GPU (Nvidia Tesla P4, 8GB) and it is a shared resource, any student can access it. Any student is allowed to use 2.5 GB at max.

But, who the fuck follows rules. Every then and now, I see some motherfucker hopping up the whole of the GPU memory. I get the username, find the guy, message him/her on Facebook to reduce the memory. Then, if I'm lucky the fucker may reduce the memory usage. Sometimes, When I lose the shit, I complain to the professor in charge too.

Most of the times, I get replies like "oh sorry, I don't know how to limit the memory usage", I'm like when the fuck will you learn to google?

Why don't people understand a shared resource is a shared resource?

Comments
  • 7
    “I Like to play with toys alone in the kindergarten”
  • 16
    That's why competent sysadmins set forced max quotas for shared resources so that this issue doesn't come up.
  • 1
    So you can see who's hogging all the resources, but you can't do anything to stop/kill whatever process they're running?
  • 1
    @endor I don't have permissions to kill anyone else's process (obviously). I usually contact the admin to warn the user for the over usage.
  • 3
    It just sounds like your institute's admin has done a bad job of managing computational resources
Your Job Suck?
Get a Better Job
Add Comment