Vanderbilt University Medical Center

Supporting Tool Suite for Production Proteomics

Cluster Queuing System


The Cluster Queuing System is a web interface that allows users to easily submit database searches using tools such as Sequest, Myrimatch, and TagRecon. It basically acts as the “middle man” between users and the distributed computing cluster. It will take the user input and submit it to be searched on the cluster, monitor the job during processing, and inform the user when the search has completed successfully or if there are errors. All of these details are logged to an Oracle database which helps track usage and parameters from historical runs.



· Application Server—Redhat Linux 5 or higher; Apache 2.2.x; Perl 5.8.x; Perl modules: DBI, DBD Pg, DBD Oracle, Compress::Raw::Bzip, Compress::Raw::Zlib, Compress::Zlib, Archive::Zip, File::Slurp, HTML::Tagset, HTML::Parser, Digest::MD5, URI, LWP, XML::Writer, XML::Parser, XML::XPath, mod_perl; Oracle client 10.2 or free Oracle Express Edition; PostgresSQL client 8.1.x; Java 1.4.2.x or above; Ant 1.8.1 or above; jsch-0.1.44.jar (for ssh ant tasks); SVN client 1.4.x or above; SVNAnt 1.2.1 or above; sendmail is configured and enabled; unix userid for appUser is already set up

· Cluster Gateway—Java JRE 1.4+; Perl 5.8.x


Installation Instructions


Database Server:

· Create Oracle database using provided script.

· Run follow-up SQL scripts on DB.

Application Server:

· Edit deploy_appserver.xml to bypass Subversion (if necessary)

· Create $HOME/deploy directory

· Copy or check out appserver/deploy/deploy_appserver.xml to $HOME/deploy directory

· Copy or check out appserver/deploy/ to $HOME/deploy directory

· Edit for your custom configuration settings

· Run the deployment using Ant: ant -f deploy_appserver.xml -propertyfile deploy-local

· Review <appHome>/conf/<cluster_username>_vhost.conf file and put in the directory where Apache looks for includes for its configuration directives.

· Review .bashrc and .cshrc (for <cluster_username>) for correct paths for $PERL5LIB and $CLUSTERAPP_HOME,

· Start or restart Apache

· Start the monitor daemon for Pipeline Queuing System: $CLUSTERAPP_HOME/monitor/

Cluster Environment:

· Copy all files in accre/bin to $HOME/bin on the Cluster computing environment

· Copy all files in accre/scripts to $HOME/scripts on the Cluster computing environment

· Copy all files in accre/tagrecon to $HOME/tagrecon on the Cluster computing environment

· Copy all files in accre/unzip60 to $HOME/unzip60 on the Cluster computing environment



Full installation details can be obtained in the Cluster_Distro directory in the distribution package or here, including

· Step-by-step installation instructions

· Full list of pre-requisite software components

· Descriptions of deployment properties