Global communication analysis and optimization
PLDI '96 Proceedings of the ACM SIGPLAN 1996 conference on Programming language design and implementation
A global communication optimization technique based on data-flow analysis and linear algebra
ACM Transactions on Programming Languages and Systems (TOPLAS)
Minimizing Data and Synchronization Costs in One-Way Communication
IEEE Transactions on Parallel and Distributed Systems
Global optimization techniques for automatic parallelization of hybrid applications
ICS '01 Proceedings of the 15th international conference on Supercomputing
Static Single Assignment Form for Message-Passing Programs
International Journal of Parallel Programming
A framework for global communication analysis of optimizations
Compiler optimizations for scalable parallel systems
Accurate Data and Context Management in Message-Passing Programs
LCPC '99 Proceedings of the 12th International Workshop on Languages and Compilers for Parallel Computing
Towards automatic translation of OpenMP to MPI
Proceedings of the 19th annual international conference on Supercomputing
An Approach To Data Distributions in Chapel
International Journal of High Performance Computing Applications
Maintaining Data Dependencies Across BPEL Process Fragments
ICSOC '07 Proceedings of the 5th international conference on Service-Oriented Computing
Slicing based code parallelization for minimizing inter-processor communication
CASES '09 Proceedings of the 2009 international conference on Compilers, architecture, and synthesis for embedded systems
| Hi-index | 0.00 |
Reducing communication overhead is crucial for improving the performance of programs on distributed-memory machines. Compilers for data-parallel languages must perform communication optimizations in order to minimize this overhead. In this paper, we show how to combine dependence analysis, traditionally used to optimize regular communication, and a data-flow analysis method originally developed to improve placement of irregular communication. Our approach allows us to perform more extensive optimizations message vectorization, elimination of redundant messages, and overlapping communication with computation. We also present preliminary experimental results that demonstrate the benefits of the proposed method.