As a part of our mission we are designing easy-to-use web interfaces for various computational biology tools. This page is a part of our new integrated system providing unified web interface for all the tools hosted by CBSU. The programs are run on CBSU computer clusters at the Cornell Theory Center. Currently there is total of 512 local processors grouped in three clusters.

Some of the tools are freely available for all, some, due to a very high computational demand, are available to registered users only. Current user id, together with user-specific functions, is displayed in the bar above.

In February 2006 CBSU was chosen to become one of ten Microsoft High-Performance Computing Institutes worldwide. As Microsoft HPC Institute CBSU participates in testing and development of new Microsoft HPC software and system solutions like Windows Compute Cluster Server (CCS) . Current version of CCS (1.0 SP1) running on CBSU's newest computer cluster (cbsum) is already linked to some of our web-based applications. P-BLAST, P-HMMER, P-IPRSCAN, MrBayes, MDIV and more are currently available on the new cluster. You can select this cluster by choosing "cbsum" in "Cluster" pull-down menus. MrBayes and MDIV are publicly available to all users, but P-BLAST, P-HMMER and P-IPRSCAN are for registered users only.

New!The Web Computing Interface along with its source code, scripts and applications is being developed together as Computational Biology Applications Suite for High Performance Computing (BioHPC Suite). This suite will be available for download along with the interface source code and will run/interface with any Microsoft-based cluster. Target date for its release is Fall 2007. You may find more details on the BioHPC Suite home page, our blog at Microsoft HPC site, and in our Plant and Animal Genome (PAG XIV) presentation or ABRF 2007 poster.

Remote Microsoft cluster available! Microsoft gave us access to their own 64 node cluster in their headquarters running MS Windows 2003 CCS, the cluster's name is Athena and it is being accessed via JSDL protocol. Currently we can fill up to 32 nodes, each dual CPU Intel Core Duo, which for most applications is equivalent to 4 CPU per node. Most of the applications are registered to run on Athena, you can check registration of any application by accessing "Cluster" pull-down menu on the application page. The inteface will automatically allocate appropriate jobs to Athena based on nodes availability, it is also possible to choose Athena from menu "Cluster". Please rememeber that Athena is a remote cluster, in order to view a file associated with a job it must be first transferred to our server, which may cause a small delay during job execution. After the job is finished all the files are transferred to our server and can be assessed in a usual way. Enjoy!

We spent a lot of time and effort developing this interface, and we hope it is a useful resource. You can help us in this effort by acknowledging our contribution to your work and research. Please acknowledge us in all publications and presentation of work that used our resources using the following text. Please send us information about your publications and presentations to the address below. Thanks!

Please contact us with any comments or questions at cbsu@tc.cornell.edu . You may visit our home page for more information about the unit.