MPI-aware compiler optimizations for improving communication-computation overlap
Anthony Danalis, Lori Pollock, Martin Swany, John Cavazos
Abstract
Anthony Danalis, Lori Pollock, Martin Swany, John Cavazos
Abstract
Several existing compiler transformations can help improve communication-computation overlap in MPI applications. However, traditional compilers treat calls to the MPI library as a black box with unknown side effects and thus miss potential optimizations. This paper's contributions enable the development of an MPI-aware optimizing compiler that can perform transformations exploiting knowledge of MPI call effects to increase communication-computa-tion overlap. We formulate a set of data flow equations and rules to describe the side effects of key MPI functions so an MPI-aware compiler can automatically assess the safety of transformations. After categorizing existing compiler transformations based on their effect on the application code, we present an optimization algorithm that specifies when and how to apply these optimizing transformations to achieve improved communication-computation overlap. By manually applying the optimization algorithm to kernels extracted from HYCOM and the NAS benchmarks, we show that even when transforming these highly optimized codes, execution time can be decreased by an average of over 30%.
OpenAlex reports 44 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
Several existing compiler transformations can help improve communication-computation overlap in MPI applications. However, traditional compilers treat calls to the MPI library as a black box with unknown side effects and thus miss potential optimizations. This paper's contributions enable the development of an MPI-aware optimizing compiler that can perform transformations exploiting knowledge of MPI call effects to increase communication-computa-tion overlap. We formulate a set of data flow equations and rules to describe the side effects of key MPI functions so an MPI-aware compiler can automatically assess the safety of transformations. After categorizing existing compiler transformations based on their effect on the application code, we present an optimization algorithm that specifies when and how to apply these optimizing transformations to achieve improved communication-computation overlap. By manually applying the optimization algorithm to kernels extracted from HYCOM and the NAS benchmarks, we show that even when transforming these highly optimized codes, execution time can be decreased by an average of over 30%.
Key concepts: Compiler, Computer science, Optimizing compiler, Interprocedural optimization, Parallel computing, Computation, Loop optimization, Set (abstract data type)