The Birch Linux Cluster is a 32-node distributed-memory multi-processor system. Each node of the cluster contains two 2.4 GHz Intel Pentium IV processors with 256KB of cache (per cpu) and 2 GB of RAM per node. The nodes are interconected with Gigabit Ethernet (60-110 Mbytes/sec bandwidth, 50-200 usecs latency) and Myricom (up to 250 Mbytes/sec bandwidth and 9 usecs or less latency). The Birch Linux cluster uses Red Hat Linux as its operating system. Tthe Portable Batch System (PBS) software is deployed to distribute the computational workload across the nodes. PBS is a batch job scheduling application that provides the facility for building, submitting and processing parallel batch jobs on the cluster.
The PBS User's Guide is online here
In order to encourage parallel use, jobs on Birch must use a minimum of 2 cpus (one node).
Obtaining an account
on the Birch Linux Cluster
Information about and criteria for obtaining a research computing enabled
account to access the Birch Linux Cluster. If you have already have a
research computing enabled account, you should proceed to the tutorial below.
Tutorial for New Users
A tutorial on running programs on the Birch Linux cluster using the
Portable Batch System (PBS) resource management software.
Birch Cluster Status
Ganglia Toolkit for displaying cluster load and status information.
Birch Queue Status
Displays jobs in the queue. Click on the link for a given job ID to get
complete information about that job.
PBS Documentation
Available Software
Intel Compilers
Intel Fortran 77/90 and C/C++ compilers for Linux.MPI
The Message Passing Interface Libraries for parallel computing.TotalView Debugger
A graphical debugger for both serial and parallel applications written in C/C++ and Fortran 77/90.IMSL F90 MP Libraries
The IMSL numerical libraries are a collection of over 1000 Fortran (F90 and F77) routines that implement algorithms useful in mathematical and statistical analysis.Research Computing Software Environments
ITC licenses the Linux version of most of its mathematical and statistical software environments including Matlab, Mathematica, Maple, IDL, SAS, and S-Plus.
Related Links
The Beowulf Project