基于P2HP平台的蛋白质折叠分布式计算

通过对基于格点模型的PERM算法进行定性分析，提出并行化计算PERM算法的策略。该策略保存了PERM算法中最关键的平均权重向量，并分发至多个计算单元进行模拟运算，从而达到减少计算时间、提高运算结果精确度的目的。以P2HP为计算平台，对多条蛋白质序列进行模拟运算。实验结果表明，基于PERM的蛋白质折叠并行计算方法能够在P2HP平台上获得较优的性能。

Distributed Computing of Protein Folding Based on P2HP

By analyzing the algorithm of PERM based on the lattice model, a strategy of how to parallelize PERM is introduced. The strategy stores the average weight vector, the most important part in PERM, as the middle information and distributes this information to several computing cells for some simulation computing. So it can improve the speed of computing and achieve the more accurate results. P2P-based High Performance Computing Platform(P2HP) is employed as the experimental environment. Some different protein sequences are computed. The results of experiment show the parallel computing method based on PERM can achieve the better performance on P2HP.