Name
Abstract | ||
|---|---|---|
When computing the makespan of a DAG, it is typically assumed that the tasks scheduled on the same computing node run in sequence. In reality, however, the tasks may be run in the time sharing manner. Our studies show that the discrepancy between the assumption of sequential execution and the reality of time sharing execution may lead to inaccurate calculation of the DAG makespan. In this paper, we first investigate the impact of the time sharing execution on the DAG makespan, and propose the method to model and determine the makespan with the time-sharing execution. Based on this model, we further develop the scheduling strategies for DAG jobs running in time-sharing. Extensive experiments have been conducted to verify the effectiveness of the proposed methods. The experimental results show that by taking time sharing into account, our DAG scheduling strategy can reduce the makespan significantly, comparing with its counterpart in sequential execution. |
Year | Venue | Field |
|---|---|---|
2018 | ICA3PP | Job shop scheduling,Computer science,Scheduling (computing),Parallel computing,Time-sharing,Distributed computing |
DocType | Citations | PageRank |
Conference | 0 | 0.34 |
References | Authors | |
18 | 6 | |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Shenyuan Ren | 1 | 0 | 1.69 |
| Ligang He | 2 | 542 | 56.73 |
| Junyu Li | 3 | 0 | 1.35 |
| Chao Chen | 4 | 2032 | 185.26 |
| Zhuoer Gu | 5 | 7 | 1.89 |
| Zhiyan Chen | 6 | 0 | 1.69 |