This paper examines load sharing in a network of workstations (NOW). It proposes a special load sharing method referred to as epoch load sharing. With this policy, load is evenly distributed among workstations with job migration only at the end of predefined intervals. The time interval between successive load sharing is called an epoch. We compare the performance of epoch load sharing with other traditional load sharing methods. The objective is to reduce the number of times that global system information is needed to make allocation decisions, while at the same time achieving good overall performance. A simulation model is used to address performance issues associated with load sharing. Simulated results indicate that epoch load sharing frequently succeeds in this pursuit.
Helen D. Karatza, Ralph C. Hilzer Jr.