Scaling data race detection for partitioned global address space programs

Park, Chang Seo; Sen, Koushik; Iancu, Costin

doi:10.1145/2464996.2465000

Cited by 9 publications

(3 citation statements)

References 40 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For example, Park et al [27] describe a precise data race detector using dynamic analysis for PGAS programs with one-sided communication, which incurs 50% overhead at 2000 cores. Their technique is directly applicable here, but its runtime overhead is higher than the total time spent by NWChem in barriers (up to 20%) in our experiments.…”

Section: Reasoning About Barrier Elisionmentioning

confidence: 99%

Barrier elision for production parallel programs

et al. 2015

Self Cite

View full text Add to dashboard Cite

Large scientific code bases are often composed of several layers of runtime libraries, implemented in multiple programming languages. In such situation, programmers often choose conservative synchronization patterns leading to suboptimal performance. In this paper, we present context-sensitive dynamic optimizations that elide barriers redundant during the program execution. In our technique, we perform data race detection alongside the program to identify redundant barriers in their calling contexts; after an initial learning, we start eliding all future instances of barriers occurring in the same calling context. We present an automatic on-the-fly optimization and a multi-pass guided optimization. We apply our techniques to NWChem-a 6 million line computational chemistry code written in C/C++/Fortran that uses several runtime libraries such as Global Arrays, ComEx, DMAPP, and MPI. Our technique elides a surprisingly high fraction of barriers (as many as 63%) in production runs. This redundancy elimination translates to application speedups as high as 14% on 2048 cores. Our techniques also provided valuable insight about the application behavior, later used by NWChem developers. Overall, we demonstrate the value of holistic context-sensitive analyses that consider the domain science in conjunction with the associated runtime software stack.

show abstract

Section: Reasoning About Barrier Elisionmentioning

confidence: 99%

Barrier elision for production parallel programs

et al. 2015

Self Cite

View full text Add to dashboard Cite

show abstract

“…Bug detection for other one-sided models. UPC-Thrille is a tool for data race detection in Unified Parallel C (UPC) programs [34], [35]. UPC is a partitioned global address space parallel programming model, which uses a one-sided communication runtime system.…”

Section: Bug Detection and Validation For General Mpi Programsmentioning

confidence: 99%

MC-Checker: Detecting Memory Consistency Errors in MPI One-Sided Applications

Chen

Dinan

Tang

et al. 2014

SC14: International Conference for High Performance Computing, Networking, Storage and Analysis

View full text Add to dashboard Cite

One-sided communication decouples data movement and synchronization by providing support for asynchronous reads and updates of distributed shared data. While such interfaces can be extremely efficient, they also impose challenges in properly performing asynchronous accesses to shared data. This paper presents MC-Checker, a new tool that detects memory consistency errors in MPI one-sided applications. MC-Checker first performs online instrumentation and captures relevant dynamic events, such as one-sided communications and load/store operations. MC-Checker then performs analysis to detect memory consistency errors. When found, errors are reported along with useful diagnostic information. Experiments indicate that MC-Checker is effective at detecting and diagnosing memory consistency bugs in MPI one-sided applications, with low overhead, ranging from 24.6% to 71.1%, with an average of 45.2%.

show abstract

“…In [13], a SAT-based predicative method is proposed to improve the generation of the schedules for revealing the deadlock errors in MPI programs. Park et al propose in [20] a sampling based dynamic analysis method for detecting data races in hybrid parallel programming. But still, input-related errors can be missed for these methods.…”

Section: Related Workmentioning

confidence: 99%

Synchronization Error Detection of MPI Programs by Symbolic Execution

Chen

Huang

et al. 2014

2014 21st Asia-Pacific Software Engineering Conference

View full text Add to dashboard Cite

Asynchrony based overlapping of computation and communication is commonly used in MPI applications. However, this overlapping introduces synchronization errors frequently in asynchronous MPI programming. In this paper, we propose a symbolic execution based method for detecting input-related synchronization errors. The path space of an MPI program is systematically explored, and the related operations of the synchronization errors in the program are checked specifically. In addition, two optimizations are proposed to improve the efficiency. We have implemented our method as a prototype tool based on the symbolic executor Cloud9. The results of the extensive experiments indicate the effectiveness of our method.

show abstract

Scaling data race detection for partitioned global address space programs

Cited by 9 publications

References 40 publications

Barrier elision for production parallel programs

Barrier elision for production parallel programs

MC-Checker: Detecting Memory Consistency Errors in MPI One-Sided Applications

Synchronization Error Detection of MPI Programs by Symbolic Execution

Contact Info

Product

Resources

About