Development Of Advanced Computer Hardware And Software

Heart, Lung, And Blood Institute

Investigators

Linked publications & trials

Abstract

With the increased availability of parallel computer resources amenable to large scale scientific computing, it is essential to optimize the use of these resources. The primary efforts include the ongoing development of parallel computing techniques suitable for macromolecular simulation and the development of a parallel computer cluster and related software for high-efficiency simulations at low cost. Current projects include: - LoBoS: High performance computing using PC clusters, LobosV - Development of parallel QM/MM methods, Replica/Path (linear scaling methods) - Development of bioinformatics tools for PC clusters - Development and support of parallel CHARMM - Development and parallelization of the EMAP method for electron microscopy applications. - Implementation and parallelization of SGLD for efficient conformational search. The LoBoS (Lots of Boxes on Shelves), Six Lobos computer cluster has been designed and constructed using commodity PCs since 1997. They provide a greater than 10-fold improvement in price/performance when compared with the traditional supercomputer vendor's offerings. The early LoBoS systems demonstrated the effectiveness of this approach and were essential in the development of techniques and linux scripts. LoBoS IV is a substantial improvement. The LoBoS Project has produced one of the most capable computational systems at the NIH for many applications involving computational chemistry tools (http://www.lobos.nh.gov/). It has opened up a new realm of high performance computing which continues to drive the cost down while improving reliability through the use of loosely coupled clusters. The main effort in developing new bioinformatics tools for the PC cluster environment is the development of a new architecture for the integration of disparate biological databases and for the distribution of compute-intensive tasks. Clusters are uniquely suited to large database applications since each compute node has its own disk (maximizing IO) and its workload may be configured to be independent of other nodes (maximizing efficiency). As part of the development of the Replica/Path QM/MM method, we have created a novel way to parallelize these large scale quantum part of these calculations involving MPI clusters.