Please ask them what additional time they need between now and March 31. Ray Bair Chief Computational Scientist, CELS Computing, Environment, and Life Sciences Argonne National Laboratory, and The University of Chicago Sent from my iPad On Feb 7, 2011, at 5:45 PM, "John Valdes" <[email protected]> wrote:
A request for additional time. Note that this is Bob Olson's project, RAST, not Folker's project, MG-RAST. Bob initially requested 50K for FY11 which was granted in full. He's now requesting 250K more (see the reason given under the "A specific reason has been given:" heading).
John
On Mon, Jan 31, 2011 at 12:00:41PM -0600, [email protected] wrote:
Hello,
A change in allocation has been requested:
Requester: olson (Robert Olson) Project: RAST Title: Rapid Genome Annotation Technology Description: The RAST server is a genome annotation server developed by the bioinformatics group in MCS that was funded by the NIH Bioinformatics Resource Centers program. Throughout the history of the RAST server, we have made use only of the computational cluster that was funded by NIH for the purpose of hosting RAST. The configuration of this cluster was extensively hand-tuned for the purposes of running RAST. We believe the technology has reached a state of maturity such that we can develop the tools to enable the core computation to execute on a general-purpose scientific cluster like the LCRC. The real test of the success of this work will be to process real genomic data at scale. To this end we will work toward the integration of the RAST server (which provides facilities for the upload of new genomic data, the monitoring of the progress of the pipeline, and tools for online browsing and analysis of the completed annotation as well as downloads of the completed annotation in one of several export formats) with the LCRC as a computational backend. Initially this integration will be manually supervised, but we plan to investigate the technology for a more automated integration. Advances in the RAST technology have decreased the overall computational load on a per-genome basis, but the number of genomes submitted to RAST continues to rise. In 2010 we were successful at porting the core RAST computations to Fusion, but at the same time we brought the more highly optimized computation online on the core RAST server, highly reducing the requirement for the computational offload. In 2011, however, we envision the return for the need for this offload. We plan to further automate the integration of RAST and Fusion, as well as to leverage the Magellan cluster to engineer a family of computational offload facilities for RAST. The RAST project has also made effective use of the Fusion resource for other computations related to the curation of the protein families that underpin the RAST system. We perform detailed all-to-all analyses of the genomes within families of organisms (for instance, all sequenced genomes in the Brucella family of pathogenic bacteria) to determine the precise relationship between the genes in these organisms. Fusion is an excellent platform for such analyses, and we plan to continue such analyses on the other pathogen groups.
Current: undetermined amount Justification:
Requested: 250000
A specific reason has been given: We have been running a new large-scale computation to compute detailed similarities between the genomes in our database. We are using the results of this data to compute detailed correspondences between genomes in the database. This data, combined with the data we curate on the linkages between annotations and the published literature, will allow us to create highly consistent and accurate annotations across the entire genome database.
The computation thus far has consumed around 40K core hours. It is roughly 18% through the total calculation (based on sequence counts), which gives us an estimate of about 224K core hours for the entire calculation. We are requesting an additional 250K core hours in order to complete this computation with some additional overage in the event our estimates are not completely accurate.
This needs to be approved and the final allocation amount decided upon.
Thank You, The LCRC Accounts System _______________________________________________ allocations-admins mailing list [email protected] https://lists.lcrc.anl.gov/mailman/listinfo/allocations-admins
allocations-admins mailing list [email protected] https://lists.lcrc.anl.gov/mailman/listinfo/allocations-admins