Formal Analysis of Message Passing

Siegel, Stephen F.; Gopalakrishnan, Ganesh

doi:10.1007/978-3-642-18275-4_2

Cited by 20 publications

(13 citation statements)

References 38 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Analysis of Parallel Programs. Previous work discusses the verification of programs using message passing interfaces, e.g., [Siegel 2005;Siegel and Avrunin 2005;Siegel and Gopalakrishnan 2011;Michael et al 2019]. Various tools are available to do model checking for Erlang, a popular actor language [Huch 1999;Fredlund and Svensson 2007;D'Osualdo et al 2013].…”

Section: Related Workmentioning

confidence: 99%

Verifying safety and accuracy of approximate parallel programs via canonical sequentialization

Fernando

Joshi

Misailovíc

2019

Proc. ACM Program. Lang.

View full text Add to dashboard Cite

We present Parallely, a programming language and a system for verification of approximations in parallel message-passing programs. Parallely's language can express various software and hardware level approximations that reduce the computation and communication overheads at the cost of result accuracy. Parallely's safety analysis can prove the absence of deadlocks in approximate computations and its type system can ensure that approximate values do not interfere with precise values. Parallely's quantitative accuracy analysis can reason about the frequency and magnitude of error. To support such analyses, Parallely presents an approximation-aware version of canonical sequentialization, a recently proposed verification technique that generates sequential programs that capture the semantics of well-structured parallel programs (i.e., ones that satisfy a symmetric nondeterminism property). To the best of our knowledge, Parallely is the first system designed to analyze parallel approximate programs. We demonstrate the effectiveness of Parallely on eight benchmark applications from the domains of graph analytics, image processing, and numerical analysis. We also encode and study five approximation mechanisms from literature. Our implementation of Parallely automatically and efficiently proves type safety, reliability, and accuracy properties of the approximate benchmarks. CCS Concepts: • Computing methodologies → Parallel programming languages; • Software and its engineering → Formal software verification.

show abstract

Section: Related Workmentioning

confidence: 99%

Verifying safety and accuracy of approximate parallel programs via canonical sequentialization

Fernando

Joshi

Misailovíc

2019

Proc. ACM Program. Lang.

View full text Add to dashboard Cite

show abstract

“…In contrast, a large proportion of existing work has aimed to provide formal analysis and verification of concurrent systems (e.g., [42], [18], [3], [63]), as well as the formal verification of dynamic and parametrized systems and networks (e.g., [9], [1], [51], [52]), as means of providing assurances that systems operate as expected as they continue to grow in size and complexity. Similarly, recent work aiming to formally identify risks and provide solutions for safely managing the complexity in the design and operation of flight-critical systems using category theory and the idea of operads has been proposed [60].…”

Section: Other Related Workmentioning

confidence: 99%

Identifying Implicit Component Interactions in Distributed Cyber-Physical Systems

Jaskolka

Villasenor

2017

Proceedings of the 50th Hawaii International Conference on System Sciences (2017)

View full text Add to dashboard Cite

Modern distributed systems and networks, like those found in cyber-physical system domains such as critical infrastructures, contain many complex interactions among their constituent software and/or hardware components. Despite extensive testing of individual components, security vulnerabilities resulting from unintended and unforeseen component interactions (so-called implicit interactions) often remain undetected. This paper presents a method for identifying the existence of implicit interactions in designs of distributed cyber-physical systems using the algebraic modeling framework known as Communicating Concurrent Kleene Algebra (C 2 KA). Experimental results verifying the applicability of C 2 KA for identifying dependencies in system designs that would otherwise be very hard to find are also presented. More broadly, this research aims to advance the specification, design, and implementation of distributed cyber-physical systems with improved cybersecurity assurance by providing a new way of thinking about the problem of implicit interactions through the application of formal methods.

show abstract

“…The type pred a used in the datatype constructors (lines [17][18][19][20] is an abbreviation of func a bool, a function of any type to a boolean. Such a predicate is used to restrict values, encoding refinement datatypes ({x : D | p} in Figure 4) This type abbreviation is part of the Why3 standard library.…”

Section: Whytheory For Protocolsmentioning

confidence: 99%

“…The objectives of MPI verification tools are diverse and include the validation of arguments to MPI primitives as well as resource usage [24], ensuring interaction properties such as the absence of deadlocks [17,20,24], or asserting functional equivalence to sequential programs [20]. The methodologies employed are also diverse, ranging from traditional dynamic analysis up to model checking and symbolic execution.…”

Section: Tools For the Verification Of Mpi Programsmentioning

confidence: 99%

Deductive Verification of Parallel Programs Using Why3

Santos

Martins

Vasconcelos

2015

Electron. Proc. Theor. Comput. Sci.

View full text Add to dashboard Cite

The Message Passing Interface specification (MPI) defines a portable message-passing API used to program parallel computers. MPI programs manifest a number of challenges on what concerns correctness: sent and expected values in communications may not match, resulting in incorrect computations possibly leading to crashes; and programs may deadlock resulting in wasted resources. Existing tools are not completely satisfactory: model-checking does not scale with the number of processes; testing techniques wastes resources and are highly dependent on the quality of the test set. As an alternative, we present a prototype for a type-based approach to programming and verifying MPI like programs against protocols. Protocols are written in a dependent type language designed so as to capture the most common primitives in MPI, incorporating, in addition, a form of primitive recursion and collective choice. Protocols are then translated into Why3, a deductive software verification tool. Source code, in turn, is written in WhyML, the language of the Why3 platform, and checked against the protocol. Programs that pass verification are guaranteed to be communication safe and free from deadlocks. We verified several parallel programs from textbooks using our approach, and report on the outcome.Comment: In Proceedings ICE 2015, arXiv:1508.0459

show abstract

Formal Analysis of Message Passing

Cited by 20 publications

References 38 publications

Verifying safety and accuracy of approximate parallel programs via canonical sequentialization

Verifying safety and accuracy of approximate parallel programs via canonical sequentialization

Identifying Implicit Component Interactions in Distributed Cyber-Physical Systems

Deductive Verification of Parallel Programs Using Why3

Contact Info

Product

Resources

About