当前位置: X-MOL 学术J. Supercomput. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Binary quantum-inspired gravitational search algorithm-based multi-criteria scheduling for multi-processor computing systems
The Journal of Supercomputing ( IF 3.3 ) Pub Date : 2020-04-28 , DOI: 10.1007/s11227-020-03292-0
Abhijeet Singh Thakur , Tarun Biswas , Pratyay Kuila

A quantum-inspired hybrid scheduling technique is proposed for multi-processor computing systems. The proposed algorithm is a hybridization of principles of quantum mechanics (QM) and a nature-inspired intelligence, gravitational search algorithm (GSA). The principles of QM such as quantum bit, superposition and rotation gate help to design an efficient agent representation as well as intense exploration capability of GSA enhances toward better converging rate. The fitness function is designed with the aim to minimize makespan, adequate balancing of loads and proper utilization of the deployed resources during the evaluation of agents. Several standard benchmarks as well as synthetic data sets are used to analyze and validate the work. The performance improvement of the proposed algorithm is compared with recently designed algorithms like quantum genetic algorithm, particle swarm optimization-based multi-criteria scheduling, Improved-GA, GSA and Cloudy-GSA. The significance of the algorithm is tested using a hypothesis analysis of variance.

中文翻译:

基于二进制量子启发引力搜索算法的多处理器计算系统多准则调度

针对多处理器计算系统提出了一种受量子启发的混合调度技术。所提出的算法是量子力学 (QM) 原理和受自然启发的智能引力搜索算法 (GSA) 的混合。量子位元、叠加和旋转门等 QM 原理有助于设计有效的代理表示以及 GSA 的强烈探索能力提高收敛速度。适应度函数的设计目的是在代理评估期间最小化完工时间、充分平衡负载和适当利用部署的资源。几个标准基准以及合成数据集用于分析和验证工作。与最近设计的量子遗传算法、基于粒子群优化的多准则调度、Improved-GA、GSA 和 Cloudy-GSA 等算法相比,该算法的性能改进。使用方差的假设分析来测试算法的显着性。
更新日期:2020-04-28
down
wechat
bug