Loading...
Please wait, while we are loading the content...
Similar Documents
Fast Address Sequence Generation for Data-Parallel Programs Using Integer Lattices (1996)
| Content Provider | CiteSeerX |
|---|---|
| Author | Thirumalai, Ashwath Ramanujam, J. |
| Description | In data-parallel languages such as High Performance Fortran and Fortran D, arrays are mapped to processors through a two step process involving alignment followed by distribution. A compiler that generates code for each processor has to compute the sequence of local memory addresses accessed by each processor and the sequence of sends and receives for a given processor to access non-local data. In this paper, we present a novel approach based on integer lattices. The set of elements referenced can be generated by integer linear combinations of basis vectors. Our linear algorithm determines the basis vectors as a function of the mapping. Using the basis vectors, we derive a loop nest that enumerates the addresses, which are points in the lattice generated by the basis vectors. Experimental results show that our approach is better than that of a recent linear time solution to this problem. |
| File Format | |
| Language | English |
| Publisher | Springer-Verlag |
| Publisher Date | 1996-01-01 |
| Publisher Institution | IN LANGUAGES AND COMPILERS FOR PARALLEL COMPUTING, C.-H. HUANG ET AL. (EDITORS), LECTURE NOTES IN COMPUTER SCIENCE |
| Access Restriction | Open |
| Subject Keyword | Novel Approach Data-parallel Program Using Integer Lattice Local Memory Address Data-parallel Language Access Non-local Data Recent Linear Time Solution Basis Vector Step Process High Performance Fortran Linear Algorithm Integer Linear Combination Experimental Result Fast Address Sequence Generation Integer Lattice |
| Content Type | Text |
| Resource Type | Article |
