Integrating bioinformatics, distributed data management, and distributed computing for applied training in high performance computing

• DOI: 10.1145/1324302.1324311
• Publication Date: 2007-12-07T14:19:41Z
• Authors: Michael D. Kane; John A. Springer
• Keywords: Computer cluster; Paging; End-user computing; Multi-core processor

The utilization of multi-core and multi-node parallel high performance computing (HPC) systems is growing rapidly to meet computational demands in the scientific arena. For example, exponential growth genomic data has outpaced increases single CPU clock speeds by 15-fold over last 20 years, placing great value on use processing bioinformatics. Fortunately, increased demand for architectures resulted decreased costs distributed components making these more affordable organizations institutions. As HPC computer grows, so does professionals skilled implementation, administration systems. With goal training undergraduate graduate students this demand, a model module been developed implemented that integrates bioinformatics, management computing. In bioinformatics provides exposure applied as well rationale multi-processor overcome large problems. addition, parallelization explored from classic divide-and-conquer approach, perspective, which places emphasis network bandwidth disk paging detractors performance. Students participate through hands-on interactions with three different cluster types: (1) Beowulf, (2) blade servers, (3) shared memory results include exploratory student projects determine mathematical relationships between nodes, type, database size segmentation methods, (4) application (5) RAM per node, (6) bandwidth. outcome across multiple types, information technology perspectives.