Datapath merging and interconnection sharing for reconfigurable architectures

Nahri Moreano, Guido Araujo, Zhining Huang, Sharad Malik

Research output: Contribution to journalConference article

24 Citations (Scopus)

Abstract

Recent work in reconfigurable computing research has shown that a substantial performance speedup can be achieved through architectures that map the most relevant application inner-loops to a reconfigurable datapath. Any solution to this problem must be able to synthesize a datapath for each loop and to merge them together into a single reconfigurable datapath. The main contribution of this paper is a novel graph-based technique for the datapath merge problem. This approach is based on the solution of a maximum clique problem that merges datapaths one at a time. A set of experiments, using the MediaBench benchmark, shows that the proposed technique produces 24% fewer datapath interconnections than a previous solution to this problem.

Original languageEnglish (US)
Pages (from-to)38-43
Number of pages6
JournalProceedings of the International Symposium on System Synthesis
StatePublished - Dec 1 2002
Event15th International Symposium on System Synthesis - Kyoto, Japan
Duration: Oct 2 2002Oct 4 2002

Fingerprint

Reconfigurable architectures
Merging
Experiments

All Science Journal Classification (ASJC) codes

  • Hardware and Architecture

Cite this

@article{ebe9f4abd6c246848aaa1c45a74f873c,
title = "Datapath merging and interconnection sharing for reconfigurable architectures",
abstract = "Recent work in reconfigurable computing research has shown that a substantial performance speedup can be achieved through architectures that map the most relevant application inner-loops to a reconfigurable datapath. Any solution to this problem must be able to synthesize a datapath for each loop and to merge them together into a single reconfigurable datapath. The main contribution of this paper is a novel graph-based technique for the datapath merge problem. This approach is based on the solution of a maximum clique problem that merges datapaths one at a time. A set of experiments, using the MediaBench benchmark, shows that the proposed technique produces 24{\%} fewer datapath interconnections than a previous solution to this problem.",
author = "Nahri Moreano and Guido Araujo and Zhining Huang and Sharad Malik",
year = "2002",
month = "12",
day = "1",
language = "English (US)",
pages = "38--43",
journal = "Proceedings of the International Symposium on System Synthesis",
issn = "1080-1820",
publisher = "IEEE Computer Society",

}

Datapath merging and interconnection sharing for reconfigurable architectures. / Moreano, Nahri; Araujo, Guido; Huang, Zhining; Malik, Sharad.

In: Proceedings of the International Symposium on System Synthesis, 01.12.2002, p. 38-43.

Research output: Contribution to journalConference article

TY - JOUR

T1 - Datapath merging and interconnection sharing for reconfigurable architectures

AU - Moreano, Nahri

AU - Araujo, Guido

AU - Huang, Zhining

AU - Malik, Sharad

PY - 2002/12/1

Y1 - 2002/12/1

N2 - Recent work in reconfigurable computing research has shown that a substantial performance speedup can be achieved through architectures that map the most relevant application inner-loops to a reconfigurable datapath. Any solution to this problem must be able to synthesize a datapath for each loop and to merge them together into a single reconfigurable datapath. The main contribution of this paper is a novel graph-based technique for the datapath merge problem. This approach is based on the solution of a maximum clique problem that merges datapaths one at a time. A set of experiments, using the MediaBench benchmark, shows that the proposed technique produces 24% fewer datapath interconnections than a previous solution to this problem.

AB - Recent work in reconfigurable computing research has shown that a substantial performance speedup can be achieved through architectures that map the most relevant application inner-loops to a reconfigurable datapath. Any solution to this problem must be able to synthesize a datapath for each loop and to merge them together into a single reconfigurable datapath. The main contribution of this paper is a novel graph-based technique for the datapath merge problem. This approach is based on the solution of a maximum clique problem that merges datapaths one at a time. A set of experiments, using the MediaBench benchmark, shows that the proposed technique produces 24% fewer datapath interconnections than a previous solution to this problem.

UR - http://www.scopus.com/inward/record.url?scp=0036957219&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0036957219&partnerID=8YFLogxK

M3 - Conference article

SP - 38

EP - 43

JO - Proceedings of the International Symposium on System Synthesis

JF - Proceedings of the International Symposium on System Synthesis

SN - 1080-1820

ER -