Semi-clairvoyant scheduling

https://doi.org/10.1016/j.tcs.2004.05.023Get rights and content
Under an Elsevier user license
open archive

Abstract

In (Symp. Discrete Algorithms 2002, p. 762) it was shown that the obvious semi-clairvoyant generalization of the Shortest Processing Time is O(1)-competitive with respect to average stretch on a single machine. In (Symp. Discrete Algorithms 2002, p. 762) it was left as an open question whether it was possible for a semi-clairvoyant algorithm to be O(1)-competitive with respect to average flow time on a single machine. Here we settle this open question by giving a semi-clairvoyant algorithm that is O(1)-competitive with respect to average flow time on a single machine. We also show a semi-clairvoyant algorithm on parallel machines that achieves up to constant factors the best known competitive ratio for clairvoyant on-line algorithms. In some sense one might conclude from this that the QoS achievable by semi-clairvoyant algorithms is competitive with clairvoyant algorithms. We finally show that, in contrast to the clairvoyant case, no semi-clairvoyant algorithm can be simultaneously O(1)-competitive with respect to average stretch and O(1)-competitive with respect to average flow time.

Keywords

Scheduling
Average flow time
Semi-clairvoyance
Simultaneous competitiveness

Cited by (0)

1

Partially supported by the IST Programme of the EU under contract ALCOMFT, APPOL II, and by the MIUR Projects “Societa’ dell'informazione”, “Algorithms for Large Data Sets: Science and Engineering” and “Efficient algorithms for sequencing and resource allocations in wireless networks”.

2

Supported in part by NSF grant CCR-0098752, NSF grant ANIR-0123705, and a grant from the US Air Force.