Scientific Computing & Visualization | Computation

This Site

Intel Pentium III Linux Cluster

Overview
Help Information
Allocations and Accounting
File Systems
Usage policies and batch

Overview

In 2003, we added an Intel Pentium III Linux Cluster to our array of high performance computing systems. The primary goal of this Cluster is to provide additional single-processor as well as multiprocessor computing capacity to augment our existing single- and multi-processor queues on the IBM pSeries machines. In addition, this Cluster can be used as a test-bed for users who want to develop or test codes in a cluster environment.

SCV's Linux Cluster consists of 30 Intel Pentium III compute nodes. Each of these nodes has 2 shared-memory processors for a total of 60 processors. The login/interactive/development machines is cootie.bu.edu. Each Cluster node has 1 GB of memory and 2 levels of cache. The primary, or L1, cache is 16 KB while the secondary, or L2, cache has 512 KB. This Cluster has two choices of network interconnect: a 100 Megabit/second Ethernet and Myricom's 2.2 Gigabit/second Myrinet.

Users with accounts on the Cluster must use ssh to log in to cootie.bu.edu or skate.bu.edu. Passwords are shared over the Scientific Computing Facilities so if you already have an account and password on our other systems, you will have the same login and password on the Cluster.

The Cluster has been in production use since December 1, 2003, with access granted to all SCF users.

Help Information

This page provides only very basic information on the Linux Cluster. For additional information, please follow the sidebar links.

If you are experiencing system problems, please send Email to help@skate.bu.edu.

For more information or help in using or porting applications to the Cluster, please contact Kadin Tseng (kadin@bu.edu) or Doug Sondak (sondak@bu.edu).

If you have questions regarding your computer account or resource allocations, please send Email to scfacct@bu.edu.

Allocations and Accounting

Our allocations policy is outlined in our SCF Users Information document. Usage on the Cluster is charged at a rate of 0.3 SUs (Service Units) per CPU hour used, reflecting the relative performance of the Linux nodes compared to the baseline, but now retired, p690 processors.

File Systems

Users have one shared home directory for the Linux Cluster and IBM Blue Gene.

As with our other systems, home directories on the Cluster are backed up nightly. If you accidentally remove a file, you can request that it be restored by sending email to help@skate.bu.edu. Make sure to specify the name (and full path such as skate:/usr1/scv/aarondf/Temp/myfile) of the file(s) that have been deleted and the date the deletion occured.

Our file storage section has additional information on what resources are available for storing your files.

Usage policies and batch

In addition to the 30 compute nodes, cootie is available for interactive development. General interactive login sessions are only allowed on this node.

The batch system running on the Cluster is openPBS (open source Portable Batch System). At present, all jobs are submitted to a single queue. Users are responsible for setting up a batch script indicating the runtime limit and other runtime parameters. Consult the runningjobs page for more detailed information on preparing your batch script and running your job. Please refer to the sidebar links for further information on code compilation and batch executions.

On the Cluster, users are restricted to using a maximum of 24 nodes (48 processors); there is currently no limit on run-time for jobs. The number of nodes that can be used simultaneously is an aggregate, so it is fine to run 24 separate single-processor jobs (one per node) at the same time.