A formal approach to detect functionally irrelevant barriers in MPI programs.
- Mathematics and Computer Science
We examine the unsolved problem of automatically and efficiently detecting functionally irrelevant barriers in MPI programs. A functionally irrelevant barrier is a set of MPI-Barrier calls, one per MPI process, such that their removal does not alter the overall MPI communication structure of the program. Static analysis methods are incapable of solving this problem, as MPI programs can compute many quantities at runtime, including send targets, receive sources, tags, and communicators, and also can have data-dependent control flows. We offer an algorithm called Fib to solve this problem based on dynamic (runtime) analysis. Fib applies to MPI programs that employ 24 widely used two-sided MPI operations. We show that it is sufficient to detect barrier calls whose removal causes a wildcard receive statement placed before or after a barrier to now begin matching a send statement with which it did not match before. Fib determines whether a barrier becomes relevant in any interleaving of the MPI processes of a given MPI program. Since the number of interleavings can grow exponentially with the number of processes, Fib employs a sound method to drastically reduce this number, by computing only the relevant interleavings. We show that many MPI programs do not have data dependent control flows, thus making the results of Fib applicable to all the input data the program can accept.
- Research Organization:
- Argonne National Lab. (ANL), Argonne, IL (United States)
- Sponsoring Organization:
- USDOE Office of Science (SC)
- DOE Contract Number:
- DE-AC02-06CH11357
- OSTI ID:
- 1009779
- Report Number(s):
- ANL/MCS/CP-61881; TRN: US201106%%997
- Journal Information:
- Lect. Notes Comput. Sci., Vol. 5205, Issue 2008; Conference: Euro PVM/MPI 2008; Sep. 7, 2008 - Sep. 10, 2008; Dublin, Ireland
- Country of Publication:
- United States
- Language:
- ENGLISH
Similar Records
Contention-free Routing for Shift-based Communication in MPI Applications on Large-scale Infiniband Clusters
RMACXX: An Efficient High-Level C++ Interface over MPI-3 RMA