SIGN-IN

Cluster dusky.sharcnet.ca

Links System documentation in the SHARCNET Help Wiki

Manufacturer HP
Operating System CentOS 7.6
Interconnect 10G
Total processors/cores 1316
Nodes
1‑20
24 cores
2 sockets x 12 cores per socket
Xeon E5-2670v3 @ 2.3 GHz
Type: Compute
Memory: 64.0 GB
Local storage: 500 GB
27
32 cores
4 sockets x 8 cores per socket
Xeon(R) CPU E5-4620 @ 2.2 GHz
Type: Compute
Memory: 1024.0 GB
Local storage: 140 GB
28‑31
12 cores
2 sockets x 6 cores per socket
Xeon(R) CPU E5-262 @ 2.0 GHz
Type: Compute
Memory: 256.0 GB
Local storage: 140 GB
32‑39
16 cores
2 sockets x 8 cores per socket
Xeon(R) CPU E5-2630 @ 2.4 GHz
Type: Compute
Notes: 8 x NVIDIA Tesla K80 GPUs
Memory: 96.0 GB
Local storage: 1.8 TB
40‑59
24 cores
2 sockets x 12 cores per socket
Xeon(R) CPU E5-2690 v3 @ 2.6 GHz
Type: Compute
Memory: 64.0 GB
Local storage: 480 GB
60‑63
24 cores
2 sockets x 12 cores per socket
Xeon(R) CPU E5-2690 v3 @ 2.6 GHz
Type: Compute
Memory: 128.0 GB
Local storage: 480 GB
76
12 cores
2 sockets x 6 cores per socket
Xeon(R) CPU E5-2620 v3 @ 2.4 GHz
Type: Compute
Memory: 768.0 GB
Local storage: 5 TB
77
40 cores
2 sockets x 20 cores per socket
Xeon(R) Gold 6148 @ 2.4 GHz
Type: Compute
Memory: 768.0 GB
Local storage: 5 TB
Total attached storage 0 Bytes
Suitable use

Note: This system is contributed by research groups. The contributing groups have the benefit of access to the resources on a preferential basis as determined on a "best efforts" basis by the SHARCNET system administrator. Jobs submitted by contributing groups have a higher priority than others. For the policies on the contribution of systems, please refer to Contribution of Computational Assets to SHARCNET.

Software available

FDTD, GDB, GCC, MONO, OCTAVE, R, INTEL, FFTW, OPENCV, EMACS, OPENMPI, AUTODOCKVINA, LSDYNA, PYTHON, BOOST, ABAQUS, COMSOL, SPARK, PERL, NINJA, SYSTEM, LAMMPS, LLVM, BINUTILS, ESPRESSO, BLAST, SUBVERSION, GIT, ACML, NCBIC++TOOLKIT, GEANT4

Recent System Notices

Status Status Notes
Aug 20 2021, 11:20AM
(2 months ago)

The cluster is back up after transferring the head node functions to one of the compute nodes. We hope to restore the compute node once the original head node has been repaired.

Aug 19 2021, 03:25PM
(2 months ago)

The entire cluster is out-of-service because the head node has powered off and we’re unable to remotely power it on again. Staff will be on-site tomorrow morning to investigate and we hope to have it back in service then.

Aug 06 2021, 09:09AM
(3 months ago)

The login node security update has been completed and the node is available again.

Aug 05 2021, 02:45PM
(3 months ago)

The login node will be rebooted at 9am Friday August 6 2021 to apply a security update. Please save any work and log out before then. It is expected to return to service within 5 minutes.

Queued or running jobs will not be affected.

Apr 19 2021, 11:15AM
(6 months ago)

Storage server is online again and /data and /scratch are available; the server has been stable for four days so we are confident the problem has been corrected.

Sign-in to get full status history