A scalability test for parallel code
- 1 December 1995
- journal article
- Published by Wiley in Software: Practice and Experience
- Vol. 25 (12) , 1299-1314
Abstract
Code scalability, crucial on any parallel system, determines how well parallel code avoids becoming a bottleneck as its host computer is made larger. The scalability of computer code can be estimated by statistically designed experiments that empirically approximate a multivariate Taylor expansion of the code's execution response function. Each suspected code bottleneck corresponds to a first‐order term in the expansion, the coefficient for that term indicating how sensitive execution is to changes in the suspect location. However, it is the expansion coefficients for second‐order interactions between code segments and the number of processors that are fundamental to discovering which program elements impede parallel speedup. A new, unified view of these second‐order coefficients yields an informal relative scalability test of high utility in code development. Discussion proceeds through actual examples, including a straightforward illustration of the test applied to SLALOM, a complex, multiphase benchmark. A quick graphical shortcut makes the scalability test readily accessible.Keywords
This publication has 6 references indexed in Scilit:
- Synthetic‐perturbation techniques for screening shared memory programsSoftware: Practice and Experience, 1994
- Scalability of parallel algorithm-machine combinationsIEEE Transactions on Parallel and Distributed Systems, 1994
- Synthetic-perturbation tuning of MIMD programsThe Journal of Supercomputing, 1994
- Categories and Context in Scalable Execution VisualizationJournal of Parallel and Distributed Computing, 1993
- Mathematics Dictionary1992
- GprofACM SIGPLAN Notices, 1982