Titlebook: Job Scheduling Strategies for Parallel Processing; IPPS/SPDP‘98 Worksho Dror G. Feitelson,Larry Rudolph Conference proceedings 1998 Springe

新娘

引水渠

Esophagitis

A resource management architecture for metacomputing systems,in such systems is complicated by five concerns that do not typically arise in other situations: site autonomy and heterogeneous substrates at the resources, and application requirements for policy extensibility, co-allocation, and online control. We describe a resource management architecture that

浪荡子

Implementing the combination of time sharing and space sharing on AP/Linux,arallel computer, the parallel computer system needs a partitioning algorithm that can partition processors for incoming tasks. However, a typical problem for the algorithm is a blockade situation, which causes low processor utilization and slow response. To avoid such a situation, we present a Time

不如屎壳郎

Job scheduling scheme for pure space sharing among rigid jobs,re space sharing among rigid jobs have been achieved by First Come First Served (FCFS). However, FCFS has a drawback such that it can not utilize processors efficiently. This paper evaluates the performance of job scheduling schemes that are proposed to alleviate the drawback of FCFS by simulation,

贪婪性

Predicting application run times using historical information,uted in the past. The novel aspect of our work is the use of search techniques to determine those application characteristics that yield the best definition of similarity for the purpose of making predictions. We use four workloads recorded from parallel computers at Argonne National Laboratory, the

Valves

fallible

使痛苦

Improving first-come-first-serve job scheduling by gang scheduling,rategy and emphasizes the notion of fairness. Severe fragmentation is prevented by using gang scheduling which is only initiated by highly parallel jobs. Good worst-case behavior of the scheduling approach has already been proven by theoretical analysis. In this paper we show by simulation with real

LAIR

Expanding symmetric multiprocessor capability through gang scheduling, However the prevailing lack of concurrent scheduling for parallel programs precludes SMP use in addressing many large-scale problems. Tightly synchronized communications are impractical and normal time-sharing reduces the benefit of cache memory. Evidence gathered at Lawrence Livermore National Lab