A data structure for managing parallel operations
- 1 January 1994
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- Vol. 2, 33-42
- https://doi.org/10.1109/hicss.1994.323281
Abstract
The data distribution directives of High Performance Fortran (HPF) provide a high-level way of describing the location of regularly distributed data which is acted on in parallel. Parallel data is acted on by parallel computations. The distribution of parallel computations is not specified in HPF, but is left up to the compiler/spl minus/typically, the owner-computes rule is used, not always the best choice. We present a set of data structures which describe data layout, consistent both with HPF directives and with more general data optimization techniques; we also present an analogous set of data structures which describe the distribution of parallel computations. These two sets of data structures are related by a set of constraints which apply when data is being accessed locally. This provides a coherent way to represent parallel data and parallel computations on an equal basis and thereby enable a compiler to optimize both kinds of parallelism.Keywords
This publication has 4 references indexed in Scilit:
- Massively parallel data optimizationPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2003
- Data optimization: minimizing residual interprocessor data motion on SIMD machinesPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2002
- Data optimization: Allocation of arrays to reduce communication on SIMD machinesJournal of Parallel and Distributed Computing, 1990
- Compiling Fortran 8x array features for the connection machine computer systemPublished by Association for Computing Machinery (ACM) ,1988