Exploiting biased load information in direct-neighbour load balancing policies

Massively parallel architectures require distributed load balancing policie s based on local load information only. This may sometime limit their effec tiveness. The paper presents a class of direct-neighbour load balancing pol icies that aim to overcome the limits of locality by introducing biases in the load information exchanged between the system nodes. This permits load information to keep into account a more global view of the system. The effi ciency of the presented policies is evaluated in dependence of the characte ristics of the system load. Experimental results show that the transmission of biased load information provides high efficiency unless the dynamicity of the load becomes too high, in which case it is preferable to exploit non -biased load information. (C) 1999 Elsevier Science B.V. All rights reserve d.