Formal Verification of Programs That Use MPI One-Sided Communication

Pervez, Salman; Gopalakrishnan, Ganesh; Kirby, Robert M.; Thakur, Rajeev; Gropp, William

doi:10.1007/11846802_13

Cited by 22 publications

(18 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A complete list of publications from this award are [5,6,7] which have appeared, one special issue journal paper based on [6] under review by the journal STTT, one invited journal paper based on [5] under review by the journal Parallel Computing, a special issue of selected papers from the TV06 workshop [10] being guest-edited by the PI and accepted by the ENTCS journal.…”

Section: Discussionmentioning

confidence: 99%

Formal Analysis for Debugging and Performance Optimization of MPI

Gopalakrishnan

Kirby

2007

2007 IEEE International Parallel and Distributed Processing Symposium

Self Cite

View full text Add to dashboard Cite

show abstract

Section: Discussionmentioning

confidence: 99%

Formal Analysis for Debugging and Performance Optimization of MPI

Gopalakrishnan

Kirby

2007

2007 IEEE International Parallel and Distributed Processing Symposium

Self Cite

View full text Add to dashboard Cite

show abstract

“…Our current implementation is tested with PVFS2, which does not support locking. We have used the algorithms presented in [15] for file locks and [17] [14] for byte range locks implementations with PVFS2. For byte range locks, we determine the start and end offsets of the byte range accessed by each process using an existing function in ROMIO i.e.…”

Section: Methodsmentioning

confidence: 99%

“…The worst case used in the experiments is the whole file locks [15], when all writes by different processes become serial. Additionally, we have also used the byte range locks as implemented in [17] [14].…”

Section: Methodsmentioning

confidence: 99%

Conflict Detection Algorithm to Minimize Locking for MPI-IO Atomicity

Sehrish

Wang

Thakur

2009

Recent Advances in Parallel Virtual Machine and Message Passing Interface

Self Cite

View full text Add to dashboard Cite

Abstract. Many scientific applications require high-performance concurrent I/O accesses to a file by multiple processes. Those applications rely indirectly on atomic I/O capabilities in order to perform updates to structured datasets, such as those stored in HDF5 format files. Current support for atomicity in MPI-IO is provided by locking around the operations, imposing lock overhead in all situations, even though in many cases these operations are non-overlapping in the file. We propose to isolate non-overlapping accesses from overlapping ones in independent I/O cases, allowing the nonoverlapping ones to proceed without imposing lock overhead. To enable this, we have implemented an efficient conflict detection algorithm in MPI-IO using MPI file views and datatypes. We show that our conflict detection scheme incurs minimal overhead on I/O operations, making it an effective mechanism for avoiding locks when they are not needed.

show abstract

“…It is also an accurate record of one of our earliest case studies that helped us gain traction in this new and important application area of applying model checking to analyze real-world high performance computing software. A shorter version of this paper appeared in [1]. As for other tools verifying MPI programs employing one-sided communication, the Marmot tool has been extended to detect incorrect usages of MPI one-sided communication commands [42].…”

Section: Introductionmentioning

confidence: 99%

Formal methods applied to high‐performance computing software design: a case study of MPI one‐sided communication‐based locking

et al. 2009

Self Cite

View full text Add to dashboard Cite

SUMMARYThere is growing need to address the complexity of verifying the numerous concurrent protocols employed in high performance computing software. Today's approaches for verification consist of testing detailed implementations of these protocols. Unfortunately, this approach can seldom show the absence of bugs, and often results in serious bugs escaping into the deployed software. An approach called Model Checking has been demonstrated to be eminently helpful in debugging these protocols early in the software life cycle by offering the ability to represent and exhaustively analyze simplified formal protocol models. The effectiveness of model checking has yet to be adequately demonstrated in high performance computing. This paper presents a case study of a concurrent protocol that was thought to be sufficiently well tested, but proved to contain two very non-obvious deadlocks in them. These bugs were automatically detected through model checking. The protocol models in which these bugs were detected were also easy to create. Recent work in our group demonstrates that even this tedium of model creation can be eliminated by employing dynamic source-code level analysis methods. Our case study comes from the important domain of Message Passing Interface (MPI) based programming which is universally employed for simulating and predicting anything from the structural integrity of combustion chambers to the path of hurricanes. We argue that model checking must be taught as well as used widely within HPC, given this and similar success stories.

show abstract

Formal Verification of Programs That Use MPI One-Sided Communication

Cited by 22 publications

References 8 publications

Formal Analysis for Debugging and Performance Optimization of MPI

Formal Analysis for Debugging and Performance Optimization of MPI

Conflict Detection Algorithm to Minimize Locking for MPI-IO Atomicity

Formal methods applied to high‐performance computing software design: a case study of MPI one‐sided communication‐based locking

Contact Info

Product

Resources

About