SIGN-IN

Cluster brown.sharcnet.ca

Links System documentation in the SHARCNET Help Wiki

Manufacturer IBM
Operating System CentOS 6
Interconnect Gigabit Ethernet
Total processors/cores 976
Nodes
brown: 1‑80
8 cores
2 sockets x 4 cores per socket
Intel Xeon @ 2.83 GHz
Type: Compute
Memory: 16.0 GB
Local storage: 124 GB
brown: 81‑94
8 cores
2 sockets x 4 cores per socket
Intel Xeon @ 2.93 GHz
Type: Compute
Memory: 12.0 GB
Local storage: 124 GB
brown: 95‑114
8 cores
2 sockets x 4 cores per socket
Intel Xeon @ 2.93 GHz
Type: Compute
Memory: 48.0 GB
Local storage: 124 GB
brown: 115‑116
24 cores
2 sockets x 12 cores per socket
AMD Opteron @ 2.2 GHz
Type: Compute
Memory: 24.0 GB
Local storage: 160 GB
brown: 119
8 cores
2 sockets x 4 cores per socket
Intel Xeon @ 2.67 GHz
Type: Login
Memory: 12.0 GB
Local storage: None
brown: 120
8 cores
2 sockets x 4 cores per socket
Intel Xeon @ 2.83 GHz
Type: Admin
Memory: 16.0 GB
Local storage: None
Total attached storage 9.77 TB
Suitable use

Threaded and serial applications with low IO requirements for shared storage.
Note: This system is contributed by a research group. 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

GAUSSIAN, COREUTILS, NWCHEM, ADF/BAND, ECLIPSE, GIT, PETSC_SLEPC, VIM, UTIL, SIESTA, CMAKE, LAMMPS, GCC, BLAST, NCL, OPENMPI, PYTHON, NETCDF, R, MPIBLAST, PARI/GP, AMBER, OCTAVE, ACML, ORCA, CPMD, SAMTOOLS, MERCURIAL, MKL, DAR, CDF, BIOSAMTOOLS, PERL, GMP, INTEL, BOOST, HDF, FFTW, GEANT4, PGI, TEXLIVE, TINKER, OPENJDK, GNU , QD, MrBAYES, ESPRESSO, OPEN64, EMACS, MPFUN90, BINUTILS, SUBVERSION, DLPOLY, GROMACS, MATLAB, LDWRAPPER, SPRNG, BIOPERL, MPFR, VALGRIND, RLWRAP, MPFUN2015, YT, MPC, CHARM++, NAMD, PNETCDF, SUPERLU, SQ, IPM, GNUPLOT, GSL, BIOPERLRUN, ILOGCPLEX, CPAN, OPENCV, ARPACK-NG, RUBY, NIX, PROOT, GHC, VMD

Current system state details Graphs

Recent System Notices

Status Status Notes
Aug 24 2016, 02:49PM
(3 months ago)

The scratch filesystem on the Brown cluster has developed unrecoverable errors which have resulted in it becoming read-only.

As this hardware is very old, and well past warranty, replacement parts are not available, and the array will be shut down and retired.

The directory itself will still be accessible as /oldscratch/$USER on brown’s login node for the next week, and will be fully disconnected on September 1st.

If you have any data on brown’s scratch, you will have until then to retrieve it from /oldscratch/$USER (where $USER is your userid) on brown’s login node, after which it will be unavailable.

Any jobs on the Brown cluster will need to be re-written to use your global work, or /tmp for temporary file access. Remember that global work files do not expire, so you will need to remember to erase temporary files yourself if you are using that location.

Any further problems should be reported to help@sharcnet.ca

Jul 29 2016, 04:28PM
(4 months ago)

Starting yesterday afternoon, /global/b has returned to service on all Sharcnet clusters. We apologize for the extended outage, and believe that there was no loss of data. In recognition of how this outage was disruptive to 1/3 of users, affected users will be given higher inter-cluster fairshare/priority.

If you notice problems, please report via the ticket system: help@sharcnet.ca or https://www.sharcnet.ca/my/problems/submit

Jul 25 2016, 12:19PM
(4 months ago)

The Brown cluster is back online after the repair of the power feed to the server room.

Jul 20 2016, 10:04AM
(5 months ago)

Due to a power failure in the server room, brown cluster is currently unavailable

Jun 27 2016, 01:23PM
(5 months ago)

The Brown cluster is back online after the power outage this morning at Waterloo.

Any jobs that were running at the time of the outage will have been killed, and will need to be restarted.

Sign-in to get full status history