A New Heuristic Approach for Reliability Optimization of Distributed Computing

K e y w o r d s D i s t r i b u t e d computer system (DCS), K-node reliability (KNR), Capacity constraint, Heuristic, K-tree disjoint reduction. 1. I N T R O D U C T I O N D i s t r i b u t e d C o m p u t i n g Sys tems (DCS) have become increas ingly popu l a r in recent years , for t he adven t of VLSI t echno logy and low-cost microprocessors has made d i s t r i b u t e d c o m p u t i n g economica l ly p rac t i ca l in t o d a y ' s c o m p u t i n g env i ronment . T h e DCS provides po t en t i a l increases in re l iabi l i ty , t h r o u g h p u t , faul t to lerance , resource shar ing, and e x t e nd ib i l i t y [1,2]. Ach iev ing increases in these pe r fo rmance charac te r i s t i cs , however, requires careful design to increase t he r e l i ab i l i ty of a DCS. A DCS has been defined as a col lect ion of nodes a t which reside c o m p u t i n g resources t h a t c o m m u n i c a t e w i th each o the r v ia a set of l inks [3]. Large scale DCSs are coming into use p r i m a r i l y because of t he economy achieved t h rough resource shar ing [4]. T h e ma in ob jec t ive of a D C S is to p rov ide efficient commun ica t i on among var ious nodes in o rder to increase the i r u t i l i t y and to make the i r service avai lable to more users [5]. One of t he fundamen ta l cons ide ra t ions in des igning such sys tems is t h a t of sys t em rel iabi l i ty , which s t rong ly de pe nds on the topo log ica l l ayou t of t he c o m m u n i c a t i o n l inks [6]. Re l i ab i l i ty is a very good measure of DCS pe r fo rmance if all the needed ne twork users are to be connec ted wi th each other . Several DCS re l i ab i l i ty *This research was partially supported by the National Science Council of the Republic of China under contract NSC82-0301-E009-013. **Author to whom all correspondence should be addressed. Typeset by ¢4A4S-TEX