E. Ayguade, J. Garcia, M. Girones, M. L. Grande, and J. Labarta, "Data redistribution in an automatic data distribution tool," in Proceedings of the 8th Workshop on Languages and Compilers for Parallel Computing, Columbus, OH, Aug. 1995. Home/Search Document Details and Download
Summary Related Articles Check |
This paper is cited in the following contexts:
Data Flow Analysis Driven Dynamic Data Partitioning - Tims, Gupta, Soffa (1998) (1 citation) (Correct)
....array variable for the duration of the program s execution. The limitation of this approach is that if data interrelationship patterns vary significantly during the course of a program, some phases of the execution will incur significant communication costs. Dynamic data distribution techniques [2, 3, 4, 6, 7, 13, 15, 17] address this limitation by allowing arrays to be realigned and redistributed during execution so that more of the program s interrelationships may be resolved by the data distribution. Care must be taken, however, that the communication saved by changing the data distribution is not offset or ....
Eduard Ayguade, Jordi Garcia, Merce Girones, M. Luz Grande, and Jesus Labarta. "Data Redistribution in an Automatic Data Distribution Tool. In Proceedings of the 8th Workshop on Languages and Compilers for Parallel Computing, pages 407-- 421, Columbus, Ohio, Aug. 1995.
A Framework for Automatic Dynamic Data Mapping - Garcia, Ayguade, Labarta (1996) (4 citations) Self-citation (Ayguade Garcia Labarta) (Correct)
.... can be either static (they do not change along program execution) or dynamic (they can change between computational phases) Most current proposals solve the problem in independent steps: alignment and distribution for each computational phase [13, 9, 11, 3, 16] and their dynamic combination [5, 4, 17, 15, 2]. The alignment step tries to find appropriate alignments between all arrays in a phase, that is, to decide for each array the dimensions that will be aligned to the dimensions of another array called the template (inter dimensional alignment) and for each aligned dimension, to decide the offset ....
....distributed. The next reference pattern is between arrays A and B, and connects all nodes in column B to all nodes in column A. The weight of the edge connecting the first dimension of array B to the first dimension of array A (A[1] B[1] for shorter) is a Local Memory Access. The weight of A[1] B[2] (a) b) A B C D 2 x NtoN 2 x NtoN NtoN NtoN NtoN NtoN NtoN NtoN 1to1 A B C D loop j2 loop i2 loop j1 Figure 2. CFG with (a) data movement edges and (b) parallelism hyperedges. is a symbolic expression representing a Many T o Many data movement primitive. This means that if these ....
[Article contains additional citation context not shown here]
E. Ayguade, J. Garcia, M. Girones, M. L. Grande, and J. Labarta. Data Redistribution in an Automatic Data Distribution Tool. In C.-H. H. et al., editor, Proceedings of the 8th Annual Workshop on Languages and Compilers for Parallel Computing, Columbus, OH, August 1995.
Optimizing Communication and Data Distribution for.. - Palermo (Correct)
No context found.
E. Ayguade, J. Garcia, M. Girones, M. L. Grande, and J. Labarta, "Data redistribution in an automatic data distribution tool," in Proceedings of the 8th Workshop on Languages and Compilers for Parallel Computing, Columbus, OH, Aug. 1995.
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC