Processor Allocation for Tasks that is Robust Against Errors in Computation Time Estimates

Abstract

Abstract

Robust Processor Allocation for Independent Tasks When Dollar Cost for Processors is a Constraint

“…The current work is an extension of our earlier work in [39], where robust static mapping heuristics against errors in ETCs were derived. The tasks were mapped to a given, fixed set of machines in [39] and dollar cost was not a constraint in that environment.…”

“…The tasks were mapped to a given, fixed set of machines in [39] and dollar cost was not a constraint in that environment. The research environment here differs from [39] with the addition of the cost constraint for the machines and choosing a subset of all the available machines to be used. Moreover, in this study, the machines are divided into classes and the machines are of consistent heterogeneity across the different classes.…”

“…Moreover, in this study, the machines are divided into classes and the machines are of consistent heterogeneity across the different classes. In [39], the machines were of inconsistent heterogeneity (i.e., the consistency property did not hold). The robustness metric used in the current work and in [39] is derived using the four step FePIA procedure detailed in [4].…”

“…The robustness metric used in the current work and in [39] is derived using the four step FePIA procedure detailed in [4]. All of the heuristics in the current paper use as a component machine assignment heuristics from our earlier work in [39].…”

“…The emphasis of this paper is on selecting the set of machines to accomplish the above stated goal. All of the heuristics in the current paper use as a component machine assignment heuristics from our earlier work in [39], which assumed a given, fixed set of machines (the earlier work did not involve selecting a set of machines for purchasing which is the focus of the current paper). Simulations are used to evaluate and compare the six static heuristics studied in this research.…”