Abstract | ||
---|---|---|
Cluster of computers are more and more popular for running applications. Such computing architectures are composed of collections of processors that are not always continuously available over time. Usually, the corresponding un-availabilities are not precisely known in advance because they are often subject to disturbances. We propose in this paper to study the impact of disturbances on the performance of the well-known LPT scheduling algorithm (Largest Processing Time) with unavailability periods. We establish new approximability bounds for this algorithm and we discuss several variants that allow to absorb the disturbances on the dates of perturbations. |
Year | DOI | Venue |
---|---|---|
2009 | 10.3233/978-1-60750-530-3-205 | PARALLEL COMPUTING: FROM MULTICORES AND GPU'S TO PETASCALE |
Keywords | Field | DocType |
list scheduling,LPT algorithm,availability constraints,uncertainties | Computer science,Scheduling (computing),Parallel computing,Unavailability,Distributed computing | Conference |
Volume | ISSN | Citations |
19 | 0927-5452 | 0 |
PageRank | References | Authors |
0.34 | 0 | 4 |
Name | Order | Citations | PageRank |
---|---|---|---|
Adel Essafi | 1 | 2 | 0.72 |
Amine Mahjoub | 2 | 8 | 1.21 |
Grégory Mounié | 3 | 128 | 9.68 |
Denis Trystram | 4 | 1120 | 160.57 |