The notion of speedup has been extensibly used in performance analysis of parallel program executions by multi-processor systems. In this paper, based on this notion, a definition of parallel degree is proposed to theoretically evaluate the parallelism of data-flow program nets, and its computation method is given. Further the fitness of the definition for the parallelism is discussed, and finally an application is suggested to estimate the number of processors required to run a given program net with a reasonable speedup.
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copy
Qi-Wei GE, Toshimasa WATANABE, Kenji ONAGA, "Analysis of Parallelism in Autonomous Execution of Data-Flow Program Nets" in IEICE TRANSACTIONS on Fundamentals,
vol. E74-A, no. 10, pp. 3008-3017, October 1991, doi: .
Abstract: The notion of speedup has been extensibly used in performance analysis of parallel program executions by multi-processor systems. In this paper, based on this notion, a definition of parallel degree is proposed to theoretically evaluate the parallelism of data-flow program nets, and its computation method is given. Further the fitness of the definition for the parallelism is discussed, and finally an application is suggested to estimate the number of processors required to run a given program net with a reasonable speedup.
URL: https://global.ieice.org/en_transactions/fundamentals/10.1587/e74-a_10_3008/_p
Copy
@ARTICLE{e74-a_10_3008,
author={Qi-Wei GE, Toshimasa WATANABE, Kenji ONAGA, },
journal={IEICE TRANSACTIONS on Fundamentals},
title={Analysis of Parallelism in Autonomous Execution of Data-Flow Program Nets},
year={1991},
volume={E74-A},
number={10},
pages={3008-3017},
abstract={The notion of speedup has been extensibly used in performance analysis of parallel program executions by multi-processor systems. In this paper, based on this notion, a definition of parallel degree is proposed to theoretically evaluate the parallelism of data-flow program nets, and its computation method is given. Further the fitness of the definition for the parallelism is discussed, and finally an application is suggested to estimate the number of processors required to run a given program net with a reasonable speedup.},
keywords={},
doi={},
ISSN={},
month={October},}
Copy
TY - JOUR
TI - Analysis of Parallelism in Autonomous Execution of Data-Flow Program Nets
T2 - IEICE TRANSACTIONS on Fundamentals
SP - 3008
EP - 3017
AU - Qi-Wei GE
AU - Toshimasa WATANABE
AU - Kenji ONAGA
PY - 1991
DO -
JO - IEICE TRANSACTIONS on Fundamentals
SN -
VL - E74-A
IS - 10
JA - IEICE TRANSACTIONS on Fundamentals
Y1 - October 1991
AB - The notion of speedup has been extensibly used in performance analysis of parallel program executions by multi-processor systems. In this paper, based on this notion, a definition of parallel degree is proposed to theoretically evaluate the parallelism of data-flow program nets, and its computation method is given. Further the fitness of the definition for the parallelism is discussed, and finally an application is suggested to estimate the number of processors required to run a given program net with a reasonable speedup.
ER -