We consider a fork-join system in which a fixed amount of computational resources has to be distributed among the K tasks forming the jobs. The queueing disciplines of the fork- and join- queues are First Come First Served. At each epoch, at most K tasks are in service while the others wait in the fork-queues. We propose an algorithm with a very simple implementation that allocates the computational resources in a way that aims at minimizing the join-queue lengths, and hence at reducing the expected job service time. We study its performance in saturation and under exponential service time and provide a methodology to derive the relevant performance indices. Explicit closed-form expressions for the expected response time and join-queue length are given for the cases of jobs consisting of two, three and four tasks.
CITATION STYLE
Marin, A., Rossi, S., & Sottana, M. (2018). Biased processor sharing in fork-join queues. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 11024 LNCS, pp. 273–288). Springer Verlag. https://doi.org/10.1007/978-3-319-99154-2_17
Mendeley helps you to discover research relevant for your work.