Resources
Condor On CSc Computing Facilities
Condor is developed by the Condor Team at the University of
Wisconsin-Madison (UW-Madison), and was first installed as a production system
in the UW-Madison Computer Sciences department nearly 10 years ago.
Condor is a specialized batch system for managing compute-intensive jobs. Like
most batch systems, Condor provides a queueing mechanism, scheduling policy,
priority scheme, and resource classifications. Users submit their compute jobs
to Condor, Condor puts the jobs in a queue, runs them, and then informs the
user as to the result.
Batch systems normally operate only with dedicated machines. Often termed
compute servers, these dedicated machines are typically owned by one
organization and dedicated to the sole purpose of running compute jobs. Condor
can schedule jobs on dedicated machines. But unlike traditional batch systems,
Condor is also designed to effectively utilize non-dedicated machines to run
jobs. By being told to only run compute jobs on machines which are currently
not being used (no keyboard activity, no load average, no active telnet users,
etc), Condor can effectively harness otherwise idle machines throughout a pool
of machines. This is important because often times the amount of compute power
represented by the aggregate total of all the non-dedicated desktop
workstations sitting on people's desks throughout the organization is far
greater than the compute power of a dedicated central resource.
Condor has several unique capabilities at its disposal which are geared towards
effectively utilizing non-dedicated resources that are not owned or managed by
a centralized resource. These include transparent process checkpoint and
migration, remote system calls, and ClassAds. Documentation on Condor can be
found at http://www.cs.wisc.edu/condor/manual.
The cy machines (cy01 -- cy05), the
Ubuntu machines in GS228 (ub228-01 -- ub228-31),
and the RA and TA desktop machines
are now running Condor.
To use it, put /opt/condor/bin in your search path. This directory
contains the executables that are used to compile and submit programs to
Condor.
The directory '/opt/condor/examples' contains example programs written
in C, C++, Fortran, and sh. A 'README' file in that directory indicates how to
compile and run them. A shell script in that directory called 'submit' will
submit these programs so as to be run by Condor.
Not that Condor jobs will
*stop* if you open a terminal window and type something into it.
Condor's main function is to make unused computing cycles available to
department users. It should be noted that the owner (primary user) of a
workstation has priority on its use. For administrative purposes additional
people may be given accounts on the workstation, if you're going to do heavy
work on someone else's personal workstation without scheduling it through
Condor, you should coordinate with them first.
Last updated September 4, 2013, by Tom Lowry
Send questions about this page to