2009
DOI: 10.1007/978-3-642-05089-3_46
|View full text |Cite
|
Sign up to set email alerts
|

Reduced Execution Semantics of MPI: From Theory to Practice

Abstract: Abstract. There is growing need to develop formal verification tools for Message Passing Interface (MPI) programs to eliminate bugs such as deadlocks and local assertion violations. Of all approaches, dynamic verification is most practical for MPI. Since the number of interleavings of concurrent programs grow exponentially, we devise a dynamic interleaving reduction algorithm (dynamic partial order reduction, DPOR) tailor-made for MPI, called POE. The key contributions of this paper are: (i) a formal semantics… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1

Citation Types

0
3
0

Year Published

2010
2010
2017
2017

Publication Types

Select...
4
3

Relationship

2
5

Authors

Journals

citations
Cited by 11 publications
(3 citation statements)
references
References 21 publications
0
3
0
Order By: Relevance
“…We presented in [14] a formal specication of such communications and determined sucient conditions under which two given communications are independent. Working only at the level of P2P communications is much simpler than the previous attempt to model the semantic of the whole MPI standard, which relied on almost 200 pages of TLA+ specication [24]. Indeed, McSimGrid would constitute a very convenient environment to study the semantics of collective algorithms, such as the non-blocking collectives.…”
Section: Dynamic Partial Ordering Reductionmentioning
confidence: 99%
“…We presented in [14] a formal specication of such communications and determined sucient conditions under which two given communications are independent. Working only at the level of P2P communications is much simpler than the previous attempt to model the semantic of the whole MPI standard, which relied on almost 200 pages of TLA+ specication [24]. Indeed, McSimGrid would constitute a very convenient environment to study the semantics of collective algorithms, such as the non-blocking collectives.…”
Section: Dynamic Partial Ordering Reductionmentioning
confidence: 99%
“…In order to detect all schedulingrelated bugs, the framework under which MPI programs are debugged needs to have the ability to determine and enforce all relevant schedules (the concept of relevant scheduled will be explained in 4.1.2). ISP (In-Situ Partial Order) [17], [18], [20], [21], the current state of the art dynamic verifier for MPI programs, is currently the only known tool that has this ability.…”
Section: Introductionmentioning
confidence: 99%
“…Our previous work achieves non-deterministic behavior coverage through an active testing or dynamic formal verification tool called ISP [4], [5], [6], [7], [8]. ISP uses the MPI profiling interface [9] to intercept MPI operations and to enforce particular outcomes for non-deterministic operations.…”
Section: Introductionmentioning
confidence: 99%