Technical Report TR607:
Yu Ma, Randall Bramley, Sun Kim
A Data Management Architecture for Computational Biology
(Jan 2005), 7 pages
Abstract: With increasing availability of high-performance computer clusters, a key bottleneck in computational biology research is the handling of metadata: information about computational jobs, locations of files, staging input and output data where it is needed, and application-specific data allowing searching and querying the results of computations. By examining the needs of PLATCOM, a gene similarity platform, general requirements are derived for a data management architecture, from which particular data management systems can be quickly composed. The prototype architecture is described and its application to genome comparative analysis in PLATCOM is shown in detail.
- Available as: