Learn More
We explore issues related to the application of finite-state verification techniques to scientific computation software employing the widely-used Message-Passing Interface (MPI). Many of the features of MPI that are important for programmers present significant difficulties for model checking. In this paper, we examine a small parallel program that computes(More)
Standard testing and debugging techniques are notoriously ineffective when applied to parallel programs, due to the numerous sources of nondeterminism arising from parallelism. MPI-Spin, an extension of the model checker Spin for verifying and debugging MPI-based parallel programs, overcomes many of the limitations associated with the standard techniques.(More)
We present a method to verify the correctness of parallel programs that perform complex numerical computations, including computations involving floating-point arithmetic. The method requires that a sequential version of the program be provided, to serve as the specification for the parallel one. The key idea is to use model checking, together with symbolic(More)
We present a method to verify the correctness of parallel programs that perform complex numerical computations, including computations involving floating-point arithmetic. This method requires that a sequential version of the program be provided, to serve as the specification for the parallel one. The key idea is to use model checking, together with(More)
This paper explores a way to apply model checking techniques to parallel programs that use the nonblocking primitives of the Message Passing Interface (MPI). The method has been implemented as an extension to the model checker Spin called Mpi-Spin. It has been applied to 17 examples from a widely-used textbook on MPI. Many correct-ness properties of these(More)
We are concerned with the verification of certain properties, such as freedom from deadlock, for parallel programs that are written using the Message Passing Interface (MPI). It is known that for MPI programs containing no " wildcard receives " (and restricted to a certain subset of MPI) freedom from deadlock can be established by considering only(More)