cluster top
By joe
- 3 minutes read - 467 wordsWell, I wrote a cluster top a while ago, and just installed it for a customer whose cluster we are burning in right now. This is an office cluster … 48 cores, has to be pretty darned silent, as it is going into an office environment. User needs to see whats running on the cluster. Top is a great interface to this. ctop is getting better.
ctop v0.25: by Scalable Informatics http://www.scalableinformatics.com
Load average: 42.12, 40.53, 30.65
Tasks: 933 total, 45 running, 888 sleeping, 0 stopped, 0 zombie
Mem 78.713 GiB total, 6.427 GiB used, 72.286 GiB free
Swap 80.013 GiB total, 0.000 GiB used, 80.013 GiB free
node
Node.pid User Vmem Rmem Shmem St %CPU %MEM Time Command
c1-2.3658 scalable 215m 56m 3396 R 100.8 0.4 21:36.07 gamess.00.x
c1-2.3659 scalable 215m 56m 3396 R 100.8 0.4 20:53.83 gamess.00.x
c1-2.3660 scalable 215m 56m 3396 R 100.8 0.4 21:01.32 gamess.00.x
c1-2.3661 scalable 215m 56m 3396 R 100.8 0.4 21:34.75 gamess.00.x
c1-1.4557 scalable 215m 56m 3396 R 100.8 0.4 20:54.71 gamess.00.x
c1-1.4559 scalable 215m 56m 3396 R 100.8 0.4 21:37.75 gamess.00.x
c1-1.4561 scalable 215m 56m 3396 R 100.8 0.4 20:52.87 gamess.00.x
c1-1.4562 scalable 215m 56m 3396 R 100.8 0.4 21:35.58 gamess.00.x
c1-1.4563 scalable 215m 56m 3396 R 100.8 0.4 21:34.97 gamess.00.x
c1-5.22800 scalable 215m 58m 3492 R 100.8 0.4 21:39.79 gamess.00.x
c1-5.22801 scalable 215m 56m 3396 R 100.8 0.4 21:34.81 gamess.00.x
c1-5.22803 scalable 215m 56m 3396 R 100.8 0.4 21:21.36 gamess.00.x
c1-5.22804 scalable 215m 56m 3396 R 100.8 0.4 21:29.27 gamess.00.x
c1-5.22805 scalable 215m 56m 3396 R 100.8 0.4 21:24.06 gamess.00.x
c1-5.22806 scalable 215m 56m 3396 R 100.8 0.4 21:24.26 gamess.00.x
c1-5.22807 scalable 215m 56m 3396 R 100.8 0.4 21:35.82 gamess.00.x
c1-3.3659 scalable 215m 56m 3396 R 100.8 0.4 21:34.05 gamess.00.x
c1-3.3662 scalable 215m 56m 3396 R 100.8 0.4 21:32.65 gamess.00.x
c1-3.3663 scalable 215m 56m 3396 R 100.8 0.4 21:33.01 gamess.00.x
c1-4.22352 scalable 215m 58m 3492 R 100.8 0.4 21:39.48 gamess.00.x
c1-4.22353 scalable 215m 56m 3396 R 100.8 0.4 21:37.51 gamess.00.x
c1-4.22355 scalable 215m 56m 3396 R 100.8 0.4 21:36.24 gamess.00.x
c1-4.22357 scalable 215m 56m 3396 R 100.8 0.4 21:35.78 gamess.00.x
c1-4.22359 scalable 215m 56m 3396 R 100.8 0.4 21:35.52 gamess.00.x
c1-2.3656 scalable 215m 58m 3492 R 98.8 0.4 21:00.64 gamess.00.x
c1-2.3657 scalable 215m 56m 3396 R 98.8 0.4 20:56.24 gamess.00.x
c1-2.3662 scalable 215m 56m 3396 R 98.8 0.4 20:56.01 gamess.00.x
c1-2.3663 scalable 215m 56m 3396 R 98.8 0.4 21:34.12 gamess.00.x
c1-1.4558 scalable 215m 56m 3396 R 98.8 0.4 21:00.46 gamess.00.x
c1-1.4560 scalable 215m 56m 3396 R 98.8 0.4 20:57.67 gamess.00.x
c1-5.22802 scalable 215m 56m 3396 R 98.8 0.4 21:25.13 gamess.00.x
Yeah, we use electronic structure codes to beat up on cores. Will do some other things to beat up on the rest of ram. I have to test MPI performance, and I have this parallel matrix multiply (mwhahaha!!!)