MPIRace-Check: Detection of Message Races in MPI Programs

Park, Mi-Young; Shim, Su Jeong; Jun, Yong-Kee; Park, Hyuk-Ro

doi:10.1007/978-3-540-72360-8_28

Cited by 15 publications

(12 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Often it is focused on identifying specific concurrency errors, e.g., to detect deadlocks [2], data races [35], message races [32], or atomicity violations [40]. We address the problem of detecting causally related event patterns that are more generic in nature and can be used to match various undesired behaviours in a system.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Towards an Efficient Online Causal-Event-Pattern-Matching Framework

Pramanik

Taylor

Wong

2013

2013 IEEE 33rd International Conference on Distributed Computing Systems

View full text Add to dashboard Cite

Event monitoring and logging, that is, recording the communication events between processes, is a critical component in many highly reliable distributed systems. The event logs enable the identification of certain safety-condition violations, such as race conditions and mutual-exclusion violations, as safety is generally contingent on processes communicating in a specific causally ordered pattern. Previous efforts at finding such patterns have often focused on offline techniques, which are unable to identify operational problems as they occur. Online monitoring tools exist but they are often restricted to identifying a specific violation condition, such as a deadlock or a race condition, using dedicated data structures. We address the more general problem of detecting causally related event patterns that can be used to identify various undesired behaviours in the system.The main challenge for online pattern matching is the need to store the partial matches to the pattern, as they may combine with future events to form a complete match. Unlike pattern matching in most other domains, causally ordered patterns can span a potentially unbounded number of events and efficiently searching through this large collection poses a significant challenge.In this paper, we introduce OCEP, an efficient online causalevent-pattern-matching framework that bounds the number of partial matches it stores by reporting only a representative subset of pattern matches. We define a subset of matches as representative if it has at least one occurrence of each event in the pattern on each process, which is applicable for a large class of distributed applications. With this definition, OCEP introduces a backtracking algorithm to efficiently find a representative subset from the history of events. An evaluation of the framework shows that OCEP is capable of handling several frequently occurring violation patterns at the event rates of some representative distributed applications.

show abstract

Section: Related Workmentioning

confidence: 99%

“…Existing methods for detecting message races pass the vector timestamp among the processes by attaching them to messages [32]. OCEP receives a vector timestamp constructed in POET, not in the application, so there is minimal extra overhead on the application itself.…”

Section: Case Studiesmentioning

confidence: 99%

Towards an Efficient Online Causal-Event-Pattern-Matching Framework

Pramanik

Taylor

Wong

2013

2013 IEEE 33rd International Conference on Distributed Computing Systems

View full text Add to dashboard Cite

show abstract

“…Afterwards, Netzer et al [11] improve their previous approaches by using a two pass hybrid on-the-fly/post-mortem scheme, and remove artifact races that are side effects of non-determinism from the bug report. In [12], Park . present an on-the-fly detection tool, which detects message races in MPI programs by checking communication concurrency in distributed processes.…”

Section: Related Workmentioning

confidence: 99%

Message Race Detection for Web Services by an SMT-Based Analysis

Elwakil

Yang

Wang

et al. 2010

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Abstract. The success of the cloud computing initiative is heavily dependent on realizing trustworthy Web Services. The trustworthiness of a Web Service is judged by four factors: security, privacy, reliability and business integrity. Web Services use message-passing for communication which opens the door for messages races. Messages race with each other when their order of arrival at a destination is not guaranteed and is affected non-deterministically by factors such as network latencies and scheduling variations. Message races are dangerous to Web Services because they can be unforeseen consequences of bugs, causing messages to arrive in an unexpected ordering. In this paper we present a novel approach for improving the reliability of Web Services by detecting message races using SMT-based analysis. We model a BPEL process as a Web Service Modeling Graph (WSMG). A WSMG model is then encoded into a set of SMT constraints. The satisfiability of these constraints means that message races will occur during the actual execution of the Web Service. Hence, we reduce the message race detection problem to constraint solving problem based on satisfiability modulo theories (SMT).

show abstract

“…There are previous tools for detecting and debugging race conditions in MPI programs such as MAD [7], MAR-MOT [4], [5], IMC [14], [15], and MPIRace-Check [11]. However, these tools are restricted to detect message races in point-to-point communication of MPI-1, and have no function to detect race conditions occurring in one-sided communication.…”

Section: Introductionmentioning

confidence: 99%

Detecting Race Conditions in One-Sided Communication of MPI Programs

Park

Chung

2009

2009 Eighth IEEE/ACIS International Conference on Computer and Information Science

Self Cite

View full text Add to dashboard Cite

While one-sided communication introduced in MPI-2 is convenient to access remote memory, it is easier for programmers to make serious errors such as race conditions. Although there are several tools for detecting race conditions in MPI programs, these tools do not provide any mechanism to detect race conditions occurring in one-sided communication. In this paper, we present a technique to detect race conditions caused by unsafe accesses of onesided operations. Our technique creates a mirror window whenever a process creates a window memory for onesided communication, and uses it to monitor and detect race conditions. Whenever any one-sided operation accesses a window memory, the corresponding mirror window will be marked and checked to see if the window memory is unsafely accessed by any concurrent operation. In our experiment, we evaluate our technique using two public benchmark programs, and the results show that our technique can precisely detect race conditions. Therefore our technique is effective to detect race conditions in MPI programs and helpful to develop reliable parallel programs.

show abstract

MPIRace-Check: Detection of Message Races in MPI Programs

Cited by 15 publications

References 8 publications

Towards an Efficient Online Causal-Event-Pattern-Matching Framework

Towards an Efficient Online Causal-Event-Pattern-Matching Framework

Message Race Detection for Web Services by an SMT-Based Analysis

Detecting Race Conditions in One-Sided Communication of MPI Programs

Contact Info

Product

Resources

About