Hybrid optimal control for load balancing in a cluster of computer nodes

Brian Moerdyk; Ray De Carlo; Douglas Birdwell; Miloš Žefran; John Chiasson

doi:10.1109/CACSD-CCA-ISIC.2006.4776899

Hybrid optimal control for load balancing in a cluster of computer nodes

Brian Moerdyk
, Ray De Carlo
, Douglas Birdwell
, Miloš Žefran
, John Chiasson

Electrical Engineering Department

Purdue University
University of Tennessee System
University of Illinois at Chicago

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

Abstract

This paper develops a hybrid optimal control technique and resulting experimental data using a previously reported deterministic dynamic nonlinear system model for load balancing in a cluster of computer nodes used for parallel computations in the presence of time delays and resource constraints. The model accounts for the trade-off between using processor resources to process tasks and transferring tasks to distribute the load evenly between the nodes to reduce overall processing time. The desired performance is achieved using hybrid optimal control techniques in a model predictive control context. Results demonstrate superior performance.

Original language	English
Pages	1713-1718
Number of pages	6
DOIs	https://doi.org/10.1109/CACSD-CCA-ISIC.2006.4776899
State	Published - 2006
Event	2006 IEEE International Conference on Control Applications, CCA 2006 - Munich, Germany Duration: 4 Oct 2006 → 6 Oct 2006

Conference

Conference	2006 IEEE International Conference on Control Applications, CCA 2006
Country/Territory	Germany
City	Munich
Period	4/10/06 → 6/10/06

Access to Document

10.1109/CACSD-CCA-ISIC.2006.4776899

Cite this

@conference{826758fc286245139c89708fe9b57eef,

title = "Hybrid optimal control for load balancing in a cluster of computer nodes",

abstract = "This paper develops a hybrid optimal control technique and resulting experimental data using a previously reported deterministic dynamic nonlinear system model for load balancing in a cluster of computer nodes used for parallel computations in the presence of time delays and resource constraints. The model accounts for the trade-off between using processor resources to process tasks and transferring tasks to distribute the load evenly between the nodes to reduce overall processing time. The desired performance is achieved using hybrid optimal control techniques in a model predictive control context. Results demonstrate superior performance.",

author = "Brian Moerdyk and \{De Carlo\}, Ray and Douglas Birdwell and Milo{\v s} {\v Z}efran and John Chiasson",

year = "2006",

doi = "10.1109/CACSD-CCA-ISIC.2006.4776899",

language = "English",

pages = "1713--1718",

note = "2006 IEEE International Conference on Control Applications, CCA 2006 ; Conference date: 04-10-2006 Through 06-10-2006",

}

TY - CONF

T1 - Hybrid optimal control for load balancing in a cluster of computer nodes

AU - Moerdyk, Brian

AU - De Carlo, Ray

AU - Birdwell, Douglas

AU - Žefran, Miloš

AU - Chiasson, John

PY - 2006

Y1 - 2006

N2 - This paper develops a hybrid optimal control technique and resulting experimental data using a previously reported deterministic dynamic nonlinear system model for load balancing in a cluster of computer nodes used for parallel computations in the presence of time delays and resource constraints. The model accounts for the trade-off between using processor resources to process tasks and transferring tasks to distribute the load evenly between the nodes to reduce overall processing time. The desired performance is achieved using hybrid optimal control techniques in a model predictive control context. Results demonstrate superior performance.

AB - This paper develops a hybrid optimal control technique and resulting experimental data using a previously reported deterministic dynamic nonlinear system model for load balancing in a cluster of computer nodes used for parallel computations in the presence of time delays and resource constraints. The model accounts for the trade-off between using processor resources to process tasks and transferring tasks to distribute the load evenly between the nodes to reduce overall processing time. The desired performance is achieved using hybrid optimal control techniques in a model predictive control context. Results demonstrate superior performance.

UR - https://www.scopus.com/pages/publications/77952891681

U2 - 10.1109/CACSD-CCA-ISIC.2006.4776899

DO - 10.1109/CACSD-CCA-ISIC.2006.4776899

M3 - Paper

AN - SCOPUS:77952891681

SP - 1713

EP - 1718

T2 - 2006 IEEE International Conference on Control Applications, CCA 2006

Y2 - 4 October 2006 through 6 October 2006

ER -

Hybrid optimal control for load balancing in a cluster of computer nodes

Abstract

Conference

Access to Document

Other files and links

Fingerprint

Cite this