论文信息 - Data Compiling for Systems of Uniform Recurrence Equations

Data Compiling for Systems of Uniform Recurrence Equations

This paper presents techniques to compile systems of recurrence equations into parallel programs defined by a set of virtual processors connected via a regular network and by the communications between these processors. These techniques are founded on a dependency analysis. The data dependencies are automatically compiled either in local memory management or in communications between the virtual processors through send/receive channels.

Catherine Mongenet | C. Mongenet

[1] J. Ramanujam,et al. Compile-Time Techniques for Data Distribution in Distributed Memory Machines , 1991, IEEE Trans. Parallel Distributed Syst..

[2] Benjamin W. Wah,et al. Systematic approaches to the design of algorithmically specified systolic arrays , 1985, ICASSP '85. IEEE International Conference on Acoustics, Speech, and Signal Processing.

[3] Erik H. D'Hollander,et al. Partitioning and Labeling of Loops by Unimodular Transformations , 1992, IEEE Trans. Parallel Distributed Syst..

[4] W. Daniel Hillis,et al. Data parallel algorithms , 1986, CACM.

[5] H. T. Kung,et al. Systolic Arrays for (VLSI). , 1978 .

[6] Catherine Mongenet,et al. Affine Timings for Systems of Affine Recurrence Equations , 1991, PARLE.

[7] Richard M. Karp,et al. The Organization of Computations for Uniform Recurrence Equations , 1967, JACM.

[8] Sanjay V. Rajopadhye. Synthesizing systolic arrays with control signals from recurrence equations , 2005, Distributed Computing.

[9] Patrice Quinton,et al. The mapping of linear recurrence equations on regular arrays , 1989, J. VLSI Signal Process..