Multi-installment load distribution in tree networks with delays

Abstract

This paper presents a new strategy for load distribution ina single-level tree network equipped with or without front-ends. The load is distributed in more than one installment in an optimal manner to minimize the processing time. This is a deviation and an improvement over earlier studies in which the load distribution is done in only one installment. Recursive equations for the general case, and their closed-form solutions for a special case in which the network has identical processors and identical links, are derived. An asymptotic analysis of the network performance with respect to the number of processors and the number of installments is carried out. Discussions of the results in terms of some practical issues like the tradeoff relationship between the number of processors and the number of installments are also presented.

Signature search time evaluation in flat file databases

“…Sending the load fraction in more than one installment so that a processor can begin its computation earlier in time will reduce the finish time and the time to start searching [3].…”

On the influence of start-up costs in scheduling divisible loads on bus networks

“…To study the ultimate performance limits in the case of single-level tree networks, a multiinstallment load distribution strategy was introduced and bounds on the performance limits have been derived [8]. Essentially, in this strategy, the processing load is distributed in more than one installment.…”

Scheduling nonlinear computational loads

“…Optimal divisible load scheduling has been developed for various interconnection topologies [14], such as linear daisy chains [6], buses [8], trees [7,15,27], hypercubes [9], and two-and three-dimensional meshes [16,17]. A number of scheduling policies have been investigated including multi-installments [18], and multi-round scheduling [11,28], simultaneous distribution [2,13] and simultaneous start [12]. Also studied are detailed parameterizations and solution time optimization [21], and combinatorial schedule optimization [19].…”