Dynamic load balancing on a distributed system

Feng, Ming-Dong; Yuen, C. K.

doi:10.1109/spdp.1994.346152

Cited by 11 publications

(2 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The majority of existing dynamic LD algorithms fall into this category [2,[5][6][7][8][9][10][11][12]. An important category of dynamic algorithms under Casavant and Kuhl's taxonomy is the dynamic suboptimal heuristic-based algorithms.…”

Section: Casavant and Kuhlmentioning

confidence: 99%

Dynamic load distribution using anti-tasks and load state vectors

Lau

Leung

1998

Concurrency: Pract. Exper.

View full text Add to dashboard Cite

Polling‐based load distribution (LD) algorithms suffer from two weaknesses: (i) load information exchanged during a polling session is confined to the two negotiating nodes only; (ii) as the distributed system grows in size (in terms of the number of constituent nodes), a larger number of polling sessions, and thus a higher amount of network bandwidth consumption and CPU overhead, are needed. We propose a new LD algorithm which is based on anti‐tasks and load state vectors. This new algorithm avoids the above weaknesses of polling‐based LD algorithms. Anti‐tasks are composite agents which travel around a distributed system to facilitate the pairing up of task senders and receivers, as well as the collection and dissemination of load information. Time‐stamped load information of processing nodes is stored in load state vectors which, when used together with anti‐tasks, encourage mutual sharing of load information among processing nodes. Anti‐tasks, which make use of load state vectors to decide their traveling paths, are spontaneously directed towards processing nodes having high transient workload, thus allowing their surplus workload to be relocated quickly. Using simulations, we evaluate the performance of our new algorithm by comparing its performance with a number of well‐known polling‐based load distribution algorithms. We found that our algorithm provides significant reduction of mean task response time over a large range of system sizes. The cost of achieving this performance gain in terms of CPU overhead and channel bandwidth consumption is generally comparable to the other algorithms we studied. © 1998 John Wiley & Sons, Ltd.

show abstract

Section: Casavant and Kuhlmentioning

confidence: 99%

Dynamic load distribution using anti-tasks and load state vectors

Lau

Leung

1998

Concurrency: Pract. Exper.

View full text Add to dashboard Cite

show abstract

“…The idea is to locate 8 queens on an 8 × 8 chessboard in mutually non-attacking positions. Since then, it has been used as a benchmark problem to demonstrate divide-and-conquer methods (Abramson and Yung, 1989), load balancing algorithms (Feng and Yuen, 1994), constraint theory (Munehisa et al, 2001) and dynamic programming (Rivin and Zabih, 1992). Parallel algorithms have also been proposed to solve the general N-queen problem.…”

Section: Introductionmentioning

confidence: 99%

An evolution search algorithm for solving N-queen problems

Gwee¹,

Lim²

2005

IJCAT

View full text Add to dashboard Cite

This paper explores evolution search algorithm for solving the N-queen problem. It will be shown how simple mechanisms of selection, reproduction and mutation can be effective in solving the N-queen problem. Simulation of the search algorithm for N up to 2000 has been achieved on a personal computer. The algorithm is robust and is capable of exploring multiple solutions to the N-queen problem. Solutions beyond the first solution uncovered are achieved without significant additional overhead. was a Lecturer in the Department of Electronic Engineering in Temasek Polytechnic, Singapore. He is presently a Faculty member of the School of EEE, Nanyang Technological University, where he is an Assistant Professor since 1999. His research interests include low power asynchronous circuit design, analog and digital Class-D amplifiers, digital signal processing, and genetic algorithms.Meng-Hiot Lim has been a Faculty member of the School of EEE, Nanyang Technological University since 1989. He received his BSc, MSc and PhD from the University of South Carolina. At present, he is an Associate Professor, with research interests in the area of computational intelligence, e-based applications, reconfigurable circuits and architecture, and financial applications of graph theory. Prior to joining the University, he has worked as a Reliability Engineer for International Rectifier, a company based in California, specialising in power MOSFETS. During his sabbatical leave from NTU in 1999/2000, he was with the Department of ECE, University of Missouri-Rolla as a Visiting Associate Professor. He is also holding a concurrent appointment as Deputy Director of the Centre for Financial Engineering, a multi-disciplinary research centre anchored at the Nanyang Business School. The Centre besides promoting multi-disciplinary research, oversees the highly regarded MSc in Financial Engineering program, where he played a significant role in the planning and formulation of the curriculum. In 1990, he initiated the School of EEE's first National Science and Technology Board (now known as A*Star) funded project. As the principal investigator of the funded project in collaboration with Seiko Instruments Inc., he worked on the development of an online fuzzy-neural diagnostic system which helped to improve on the overall productivity of critical processes in the manufacture of crystal quartz resonators. In 1996, he won first prize in the AI Challenge Trophy competition organised by the Singapore Computer Society with his winning program, an evolutionary algorithm which produced the best outcome for an operational research problem. During his time with NTU, he has also offered technical consultancy for companies and delivered specialised short courses. In 1997, he was engaged by SingTel Yellow Pages to undertake the development of a large-scale computerised resource allocation system. Within a pressing schedule of six months, he single-handedly implemented the turnkey system based on a multi-stage computational intelligence techniques tailored specifica...

show abstract

An adaptive load distribution algorithm for resolving bursty workload

Lau

1999

Concurrency: Pract. Exper.

View full text Add to dashboard Cite

Dynamic load balancing on a distributed system

Cited by 11 publications

References 6 publications

Dynamic load distribution using anti-tasks and load state vectors

Dynamic load distribution using anti-tasks and load state vectors

An evolution search algorithm for solving N-queen problems

An adaptive load distribution algorithm for resolving bursty workload

Contact Info

Product

Resources

About