Synthesizing Efficient VLSI Array Processors from Iterative Algorithms by Excluding Pseudo-Dependences

Yeong-Sheng CHEN; Sheng-De WANG; Kuo-Chun SU

Synthesizing Efficient VLSI Array Processors from Iterative Algorithms by Excluding Pseudo-Dependences

Yeong-Sheng CHEN, Sheng-De WANG, Kuo-Chun SU

Full Text Views

0

Share
Cite this

Summary :

This paper is concerned with synthesizing VLSI array processors from iterative algorithms. Our primary objective is to obtain the highest processor efficiency but not the shortest completion time. Unlike most of the previous work that assumes the index space of the given iterative algorithm to be boundless, the proposed method takes into account the effects of the boundaries of the index space. Due to this consideration, the pseudo-dependence relations are excluded, and most of the independent computations can therefore be uniformly grouped. With the method described in this paper, the index space is partitioned into equal-size blocks and the corresponding computations are systematically and uniformly mapped into processing elements. The synthesized VLSI array processors possess the attractive feature of very high processor efficiency, which, in general, is superior to what is derived from the conventional linear transformation methods.

Publication: IEICE TRANSACTIONS on Fundamentals Vol.E78-A No.10 pp.1369-1380

Publication Date: 1995/10/25

Publicized

Online ISSN

DOI

Type of Manuscript: PAPER

Category: Digital Signal Processing

Cite this

Copy

Yeong-Sheng CHEN, Sheng-De WANG, Kuo-Chun SU, "Synthesizing Efficient VLSI Array Processors from Iterative Algorithms by Excluding Pseudo-Dependences" in IEICE TRANSACTIONS on Fundamentals, vol. E78-A, no. 10, pp. 1369-1380, October 1995, doi: .
Abstract: This paper is concerned with synthesizing VLSI array processors from iterative algorithms. Our primary objective is to obtain the highest processor efficiency but not the shortest completion time. Unlike most of the previous work that assumes the index space of the given iterative algorithm to be boundless, the proposed method takes into account the effects of the boundaries of the index space. Due to this consideration, the pseudo-dependence relations are excluded, and most of the independent computations can therefore be uniformly grouped. With the method described in this paper, the index space is partitioned into equal-size blocks and the corresponding computations are systematically and uniformly mapped into processing elements. The synthesized VLSI array processors possess the attractive feature of very high processor efficiency, which, in general, is superior to what is derived from the conventional linear transformation methods.
URL: https://globals.ieice.org/en_transactions/fundamentals/10.1587/e78-a_10_1369/_p

Copy

@ARTICLE{e78-a_10_1369,
author={Yeong-Sheng CHEN, Sheng-De WANG, Kuo-Chun SU, },
journal={IEICE TRANSACTIONS on Fundamentals},
title={Synthesizing Efficient VLSI Array Processors from Iterative Algorithms by Excluding Pseudo-Dependences},
year={1995},
volume={E78-A},
number={10},
pages={1369-1380},
abstract={This paper is concerned with synthesizing VLSI array processors from iterative algorithms. Our primary objective is to obtain the highest processor efficiency but not the shortest completion time. Unlike most of the previous work that assumes the index space of the given iterative algorithm to be boundless, the proposed method takes into account the effects of the boundaries of the index space. Due to this consideration, the pseudo-dependence relations are excluded, and most of the independent computations can therefore be uniformly grouped. With the method described in this paper, the index space is partitioned into equal-size blocks and the corresponding computations are systematically and uniformly mapped into processing elements. The synthesized VLSI array processors possess the attractive feature of very high processor efficiency, which, in general, is superior to what is derived from the conventional linear transformation methods.},
keywords={},
doi={},
ISSN={},
month={October},}

Copy

TY - JOUR
TI - Synthesizing Efficient VLSI Array Processors from Iterative Algorithms by Excluding Pseudo-Dependences
T2 - IEICE TRANSACTIONS on Fundamentals
SP - 1369
EP - 1380
AU - Yeong-Sheng CHEN
AU - Sheng-De WANG
AU - Kuo-Chun SU
PY - 1995
DO -
JO - IEICE TRANSACTIONS on Fundamentals
SN -
VL - E78-A
IS - 10
JA - IEICE TRANSACTIONS on Fundamentals
Y1 - October 1995
AB - This paper is concerned with synthesizing VLSI array processors from iterative algorithms. Our primary objective is to obtain the highest processor efficiency but not the shortest completion time. Unlike most of the previous work that assumes the index space of the given iterative algorithm to be boundless, the proposed method takes into account the effects of the boundaries of the index space. Due to this consideration, the pseudo-dependence relations are excluded, and most of the independent computations can therefore be uniformly grouped. With the method described in this paper, the index space is partitioned into equal-size blocks and the corresponding computations are systematically and uniformly mapped into processing elements. The synthesized VLSI array processors possess the attractive feature of very high processor efficiency, which, in general, is superior to what is derived from the conventional linear transformation methods.
ER -