Transparent_banner
home || sign-in || register ||

Cluster requin.sharcnet.ca

Links System documentation in the SHARCNET Help Wiki

Manufacturer HP
Operating System HP Linux XC 3.2.1
Interconnect Quadrics Elan4
Total processors/cores 1546
Nodes
requin: 1‑768
2 cores
2 sockets x 1 core per socket
AMD Opteron @ 2.6 GHz
Type: Compute
Notes: N/A
Memory: 8.0 GB
Local storage: 160 GB
requin: 769‑770
2 cores
2 sockets x 1 core per socket
AMD Opteron @ 2.6 GHz
Type: Login
Notes: Login nodes.
Memory: 4.0 GB
Local storage: None
requin: 771‑773
2 cores
2 sockets x 1 core per socket
AMD Opteron @ 2.6 GHz
Type: Admin
Notes: Admin nodes.
Memory: 4.0 GB
Local storage: None
Total attached storage 68.4 TB
Suitable use

Large, tightly-coupled MPI jobs.

Software available

BLACS, PATHSCALE, GAUSSIAN, MPFR, PGI, GSL, GIT, BLAST, NWCHEM, UTIL, CHARM++, GMP, MrBAYES, OPT, MATLAB, HDF, GCC, DDT, HP-MPI, INTEL, FIREFLY, OPEN64, VIM, BINUTILS, ACML, GNUPLOT, GNU , COMPILE, FFTW, MPC, SQ, LDWRAPPER

Current system state details Graphs

Recent System Notices

Status Status Notes
Jun 14 2016, 07:15PM
(13 days ago)

Requin remains offline due to a problem with its networking. We hope to have it back online around the 28th.

Jun 12 2016, 01:36AM
(15 days ago)

Due to a datacenter cooling failure, Wobbie, Requin and Iqaluk are offline.

Mar 11 2016, 04:06PM
(4 months ago)

Requin is back in service: nodes are expected to behave as before, but /scratch will be decommissioned.

Requin’s local storage (ie, not home directories or global work) has not been recommended for some time now, and is no longer usable after the recent power outage killed 4 disks at once in a single array.

Feb 20 2016, 05:15PM
(4 months ago)

Campus facilities has tentatively required a brief outage of our chillers, which will require shutting down all systems located in the McMaster datacenter: Wobbie, Requin, and Iqaluk.

If finalized, this outage will take place Friday, February 26, between 7:30-8:30am. Systems will avoid starting jobs leading up to that time, and will be shutdown shortly before the outage. Systems should return to normal service before noon.

Dec 09 2015, 09:48PM
(7 months ago)

Jobs are running normally.

Sign-in to get full status history