Finding idle machines in a workstation-based distributed system

A physical particle and plane framework for load balancing in multiprocessors

“…Most of the proposals use a decentralized algorithm where load balancing agents at the particular nodes negotiate the possible transfer of tasks from overloaded to under-loaded nodes. Most of the dynamic load balancing algorithms reported in the literature have been put forth to operate on local area networks where all of the processors can be considered as adjacent nodes [5,7,17]. Hence, it is not strange to see that in most of the cases the proposed algorithms have ignored some of the issues like inter-process communication, resource dependency and fault-tolerance which are deemed to be critical in certain multiprocessor systems.…”

Providing structure for medium-grained distributed object-based computation

“…A number of communication harnesses have been developed to steal cycles from workstations either in a consensual fashion [19] or from within an open network [35]. These activities were opportunist and were not promoted as a general solution.…”

Finite Buffers and Fast Multicast

“…To illustrate this last point, consider the problem of finding a lighdy-loaded site for balancing. Theimer [10] suggests that sites schedule their responses proportionally to their ~oads and ?ot respond at all if they are unwilling to accept more JObs: Assrgned response instants increase the expected time to recerv~ a sm~ n~ber of responses, and is too rigid of a policy. We will consrder 1.1.d.…”