A Fault Tolerant Optimization Algorithm based on Evolutionary Computation

In this paper we describe how an evolutionary algorithm is capable of running on a distributed environment with volatile resources. When executing algorithms in a desktop computing or resource harvesting context, resources can be reclaimed by their owners without warning, which may produce data loss and process to fail. The interest of the algorithm presented in the paper is that although it doesn't keep processes from failing, or data from being lost, it does improve the quality of results because of its design, not employing any special task control, checkpoint/restart or resource redundancies. By means of a series of experiments, we test the performance of the algorithm by studying the number of process failing and the quality of solutions when compared with the classic flavor of the evolutionary algorithm. The new algorithm, which shows its advantages, therefore improve dependability of distributed system

[1]  F. Fernández,et al.  Saving Effort in Parallel GP by Means of Plagues , 2004, EuroGP.

[2]  David P. Anderson,et al.  BOINC: a system for public-resource computing and storage , 2004, Fifth IEEE/ACM International Workshop on Grid Computing.

[3]  El-Ghazali Talbi,et al.  Parallel hybrid multi-objective island model in peer-to-peer environment , 2005, 19th IEEE International Parallel and Distributed Processing Symposium.

[4]  Jon B. Weissman.  Fault tolerant computing on the grid: what are my options? , 1999, Proceedings. The Eighth International Symposium on High Performance Distributed Computing (Cat. No.99TH8469).

[5]  Felix C. Freiling,et al.  Fundamentals of Fault-Tolerant Distributed Computing in Asynchronous Environments , 1999, ACM Comput. Surv..

[6]  N. Pierce.  Origin of Species , 1914, Nature.

[7]  Christian Engelmann,et al.  Super-Scalable Algorithms for Computing on 100, 000 Processors , 2005, International Conference on Computational Science.

[8]  Francisco Fernández de Vega,et al.  Saving Resources with Plagues in Genetic Algorithms , 2004, PPSN.

[9]  Vidroha Debroy,et al.  Genetic Programming , 1998, Lecture Notes in Computer Science.

[10]  Miron Livny,et al.  Checkpoint and Migration of UNIX Processes in the Condor Distributed Processing System , 1997 .

[11]  William H. Sanders,et al.  A performability-oriented software rejuvenation framework for distributed applications , 2005, 2005 International Conference on Dependable Systems and Networks (DSN'05).

[12]  David E. Goldberg,et al.  Predicting Speedups of Ideal Bounding Cases of Parallel Genetic Algorithms , 1997, ICGA.

[13]  Zbigniew Michalewicz,et al.  Genetic Algorithms + Data Structures = Evolution Programs , 1996, Springer Berlin Heidelberg.

[14]  Douglas Thain,et al.  Building Reliable Clients and Services , 2004, The Grid 2, 2nd Edition.

[15]  Leonardo Vanneschi,et al.  An Empirical Study of Multipopulation Genetic Programming , 2003, Genetic Programming and Evolvable Machines.