Improving distributed memory applications testing by message perturbation

Vuduc, Richard; Schulz, Martin; Quinlan, Dan; Supinski, Bronis de; Saebjornsen, Andreas

doi:10.1145/1147403.1147409

Cited by 21 publications

(16 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…An idealized MPI dynamic analyzer must possess (at least) the following features: De-bias from the absolute speeds of the platform: Conventional execution based testing methods for MPI omit many important schedules, causing them to miss bugs even in short MPI programs [9]. This is mainly because of the fact that their executions get trapped into a narrow range of all feasible schedules [52]. To illustrate this issue, consider Figure 4.…”

Section: Dynamic Analysis Of Mpimentioning

confidence: 99%

Formal Analysis of Message Passing

Siegel

Gopalakrishnan

2011

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Abstract. The message passing paradigm underlies many important families of programs-for instance programs in the area of high performance computing that support science and engineering research. Unfortunately, very few formal methods researchers are involved in developing formal analysis tools and techniques for message passing programs. This paper summarizes research being done in our groups in support of this area, specifically with respect to the Message Passing Interface. We emphasize the need for specialized varieties of many familiar notions such as deadlock detection, race analysis, symmetry analysis, partial order reduction, static analysis and symbolic reasoning support. Since these issues are harbingers of those being faced in multicore programming, the time is ripe to build a critical mass of researchers working in this area.

show abstract

Section: Dynamic Analysis Of Mpimentioning

confidence: 99%

Formal Analysis of Message Passing

Siegel

Gopalakrishnan

2011

Lecture Notes in Computer Science

View full text Add to dashboard Cite

show abstract

“…A Jumpshot (or other [10]) view of an MPI program execution driven by simple test harnesses often shows the temporal clustering and recurrence of MPI communications. Such tests (and hence visualizations) rarely involve the elusive 'corner case' interleavings that lead to bugs.…”

Section: Background Related Workmentioning

confidence: 99%

How Formal Dynamic Verification Tools Facilitate Novel Concurrency Visualizations

Aananthakrishnan

DeLisi

Vakkalanka

et al. 2009

Recent Advances in Parallel Virtual Machine and Message Passing Interface

View full text Add to dashboard Cite

Abstract. With the exploding scale of concurrency, presenting valuable pieces of information collected by formal methods tools intuitively and graphically can greatly enhance concurrent system debugging. Traditional MPI program debuggers present trace views of MPI program executions. Such views are redundant, often containing equivalent traces that permute independent MPI calls. In our ISP formal dynamic verifier for MPI programs, we present a collection of alternate views made possible by the use of formal dynamic verification. Some of ISP's views help pinpoint errors, some facilitate discerning errors by eliminating redundancy, while others help understand the program better by displaying concurrent even orderings that must be respected by all MPI implementations, in the form of completes-before graphs. In this paper, we describe ISP's GUI capabilities in all these areas which are currently supported by a portable Java based GUI, a Microsoft Visual Studio GUI, and an Eclipse based GUI whose development is in progress.

show abstract

“…Indeed, a few such tools exist already. Several focus on dynamic error detection, including MPI-CHECK [21], Umpire [32], MARMOT [19], the Intel Message Checker [11], and our own work on JitterBug [33].…”

Section: Related Workmentioning

confidence: 99%

Techniques for specifying bug patterns

Quinlan

Vuduc

Misherghi

2007

Proceedings of the 2007 ACM Workshop on Parallel and Distributed Systems: Testing and Debugging

View full text Add to dashboard Cite

We present our on-going work to develop techniques for specifying source code signatures of bug patterns. Specifically, we discuss two approaches. The first approach directly analyzes a program in the intermediate representation (IR) of the ROSE compiler infrastructure using ROSE's API. The second analyzes the program using the bddbddb system of Lam, Whaley, et al.. In this approach, we store the IR produced by ROSE as a relational database, express patterns as declarative inference rules on relations in the language Datalog, and bddbddb implements the Datalog programs using binary decision diagram (BDD) techniques. Both approaches readily apply to large-scale applications, since ROSE provides full type analysis, control flow, and other available analysis information. In this paper, we primarily consider bug patterns expressed with respect to the structure of the source code or the control flow, or both. More complex techniques to specify patterns that are functions of data flow properties may be addressed by either of the above approaches, but are not directly treated here.Our Datalog-based work includes explicit support for expressing patterns on the use of the Message Passing Interface (MPI) in parallel distributed memory programs. We show examples of this on-going work as well.

show abstract

Improving distributed memory applications testing by message perturbation

Cited by 21 publications

References 14 publications

Formal Analysis of Message Passing

Formal Analysis of Message Passing

How Formal Dynamic Verification Tools Facilitate Novel Concurrency Visualizations

Techniques for specifying bug patterns

Contact Info

Product

Resources

About