Title
Optimization And Parallelization Of The Isotopic Mass-To-Charge Ratio And Envelope Fingerprinting Algorithm On Supervessel Cloud
Abstract
To accommodate the new features of modern protein mass spectra with Nobel-prize-winner electrospray ionization, Zhixin Tian, et al. developed isotopic Mass-to-charge ratio and Envelope Fingerprinting (iMEF) algorithm for in situ interpretation and database search of protein tandem mass spectra. The creation of the customized theoretical database of both proteins and their dissociated fragment ions requires efficient computation of isotopic envelopes. This paper presents an optimized parallel algorithm for rapid computation of isotopic envelopes on IBM SuperVessel Cloud Platform based on OpenPOWER, and mainly adopt pre storage strategy with IBM DB2 and parallelization based on OpenMP and MPI to implement an effective application to calculate isotopic envelopes with high performance. With optimization on pre storage, the program saves lots of time from redundant computing. And by parallelization within and among tasks, we compare the output result of performance and conclude that parallelization among tasks has better performance than the program paralleled within tasks. The speedup can achieve 31 with 90 threads on a single Power8 node. On a cluster of three nodes with MPI and OpenMP combined, the speed up can reach 86.4. The experimental results show that parallel algorithm with multiple optimization strategies provide an effective method of high performance to compute isotopic envelopes
Year
Venue
Keywords
2015
2015 8TH INTERNATIONAL CONFERENCE ON BIOMEDICAL ENGINEERING AND INFORMATICS (BMEI)
iMEF, SuperVessel, OpenPOWER, Pre Storage, DB2, Parallelization, OpenMP, MPI
Field
DocType
Citations 
Tandem,Parallel algorithm,Computer science,Effective method,Parallel computing,Algorithm,POWER8,Thread (computing),Speedup,Cloud computing,Computation
Conference
0
PageRank 
References 
Authors
0.34
1
5
Name
Order
Citations
PageRank
Jingpeng Wang101.35
Jie Huang200.34
Hao Chen331.22
Mi Li433.10
Zhixin Tian500.34