|933,044||CPU Core Hours|
Named after the waterway that cuts across the
Keweenaw Peninsula, the
and built with
Rocks Cluster Distribution 6.1.1
(with CentOS 6.3), Portage is a
shared high-performance computing cluster.
It is used for testing, educational (course work and/or senior design projects) and gateway/preliminary research projects involving non-confidential and non-sponsored data.
It has one front end, one login node, one 6 TB RAID60 NAS node serving the
partition, fivetraditional CPU compute nodes [each
having 16 CPU cores (Intel Sandy Bridge E5-2670)
and 64 GB RAM] and one GPU compute node [each
having 16 CPU cores (Intel Sandy Bridge E5-2670),
64 GB RAM and 4 NVIDIA Tesla M2090 GPUs].
A Gigabit ethernet backend network serves the administrative needs of this cluster, and a 56 Gbps InfiniBand network serves its computing needs.
Originally started by National Partnership for
Advanced Computing Infrastructure and
San Diego Supercomputer Center
and called NPACI Rocks,
Rocks Cluster Distribution
is aimed at making every aspect (deploy, manage,
upgrade and scale) of high performance computing
Rocks is supported by the National Science Foundation, under grants OCI-0721623 and OCI-1032778.
Dr. Ossama Abdelkhalik
Associate Professor, ME-EM
Dr. Abdelkhalik's research explores techniques, based on the idea of hidden genes, which enable GAs to select ... [Read more]
Dr. Timothy Havens
Associate Professor, ECE/CS
Dr. Havens's research focuses on applications of and methods in machine learning and pattern recognition ... [Read more]
- Open ServiceDesk Plus Portal
- Log in with ISO credentials
- Use a brief but descriptive ISSUE TITLE in the Subject field
Including as much information as possible (for e.g., command sequence/work flow that is causing the issue, job ID, full path to the folder where the simulation was started from, as is error messages, screenshots, etc.) will help diagnose and resolve the issue in a timely manner.