Scheduling optimization through iterative refinement

Mayez Al-Mouhamed; Adel Al-Masarani

Scheduling optimization through iterative refinement

Mayez Al-Mouhamed^*
, Adel Al-Masarani

^*Corresponding author for this work

King Fahd University of Petroleum & Minerals

Research output: Contribution to conference › Paper › peer-review

1 Scopus citations

Abstract

Scheduling DAGs with communication times is the theoretical basis for achieving efficient parallelism on distributed systems. Graham's task-level is generalized to incorporate the effects of computation, data size, and network latency. A new scheduling that uses the proposed task-level to make early reservation of resources for critical computation and communication is proposed, as well as an optimization called Iterative Refinement Scheduling that alternatively schedules the computation graph and its associated reverse.

Original language	English
Pages	178-184
Number of pages	7
State	Published - 1995

Bibliographical note

Funding Information:
The authors acknowledge computing support and conference attendence support from the King Fahd University of Petroleum and Minerals, Dhahran 31261, Saudi Arabia.

ASJC Scopus subject areas

General Engineering

Cite this

@conference{b19ca8004ba447949407b8955fff731e,

title = "Scheduling optimization through iterative refinement",

abstract = "Scheduling DAGs with communication times is the theoretical basis for achieving efficient parallelism on distributed systems. Graham's task-level is generalized to incorporate the effects of computation, data size, and network latency. A new scheduling that uses the proposed task-level to make early reservation of resources for critical computation and communication is proposed, as well as an optimization called Iterative Refinement Scheduling that alternatively schedules the computation graph and its associated reverse.",

author = "Mayez Al-Mouhamed and Adel Al-Masarani",

note = "Funding Information: The authors acknowledge computing support and conference attendence support from the King Fahd University of Petroleum and Minerals, Dhahran 31261, Saudi Arabia. ",

year = "1995",

language = "English",

pages = "178--184",

}

TY - CONF

T1 - Scheduling optimization through iterative refinement

AU - Al-Mouhamed, Mayez

AU - Al-Masarani, Adel

N1 - Funding Information: The authors acknowledge computing support and conference attendence support from the King Fahd University of Petroleum and Minerals, Dhahran 31261, Saudi Arabia.

PY - 1995

Y1 - 1995

N2 - Scheduling DAGs with communication times is the theoretical basis for achieving efficient parallelism on distributed systems. Graham's task-level is generalized to incorporate the effects of computation, data size, and network latency. A new scheduling that uses the proposed task-level to make early reservation of resources for critical computation and communication is proposed, as well as an optimization called Iterative Refinement Scheduling that alternatively schedules the computation graph and its associated reverse.

AB - Scheduling DAGs with communication times is the theoretical basis for achieving efficient parallelism on distributed systems. Graham's task-level is generalized to incorporate the effects of computation, data size, and network latency. A new scheduling that uses the proposed task-level to make early reservation of resources for critical computation and communication is proposed, as well as an optimization called Iterative Refinement Scheduling that alternatively schedules the computation graph and its associated reverse.

UR - https://www.scopus.com/pages/publications/0029179470

M3 - Paper

AN - SCOPUS:0029179470

SP - 178

EP - 184

ER -

Scheduling optimization through iterative refinement

Abstract

Bibliographical note

ASJC Scopus subject areas

Fingerprint

Cite this