Resources Allocation Optimization in Distributed and Heterogeneous Computing Environments

In this work, we introduce a slot co-allocation algorithm for parallel jobs execution in distributed computing with non-dedicated and heterogeneous resources. A single slot is a time span that can be assigned to a task, which is a part of a parallel job. The job launch requires a co-allocation of a specified number of slots starting and finishing synchronously. Some existing algorithms assign a job to the first set of slots matching the resource request without any optimization (the first fit type), while other algorithms are based on an exhaustive search. In this paper, we refine a general first fit resources selection algorithm by incorporating a special slots subset allocation procedure. The improved algorithm provides an optimal solution for a considered scheduling problem and maintains a reasonable computational complexity. The algorithm is compared with two common resources allocation heuristics and its optimization efficiency is proved by a simulation study against time, cost and general-case criteria.