Dynamic Two-Side Matching of Tasks and Resources in Wide-Area Distributed Computing Environments.
Yao Song,Liang Wang,Limin Xiao,Runnan Shen,Jinquan Wang,Chenhao Zhang
DOI: https://doi.org/10.1007/s11227-023-05056-y
IF: 3.3
2023-01-01
The Journal of Supercomputing
Abstract:Recently, wide-area distributed computing environments have become popular due to its huge resource capacities. In the wide-area distributed computing environment, matching between tasks and resources plays an important role in improving system performance. However, the geographically distribution of diverse resources complicates the matching problems, thus degrading the system performance. To achieve an efficient matching between task requirements and resource capacities, this study proposes a dynamic two-side matching of tasks and resources in wide-area distributed computing environments. First, the proposed method establishes the task requirement model and the resource capacity model using a uniform description based on the characteristic analysis to mitigate the impact of diversity and simplify the matching process. Then, a two-side matching degree metric is designed to comprehensively quantify the matching quality between the task requirements and resource capacities. Finally, a resource selection strategy is raised to guide the scheduling based on the matching degree. The experimental results indicate that compared with the state-of-the-art matching methods, the proposed method reduces the overall completion time and response delay by up to 35.60% and 29.28%, respectively.