Page 115 - 《软件学报》2021年第12期

王静莲 等:软硬件节能原理深度融合之绿色异构调度算法 3779

(a) (b)
Fig.7 Comparison of disk utilization after scheduling data intensive tasks by MaOEA/C [23] and GHSA_di/II
图 7 MaOEA/C 算法 [23] 和 GHSA_di/II 算法调度数据密集型任务后的硬盘利用率对比
4 结 论

计算云高性能向高效能演进,是人类可持续发展迫切需求;同时,减少广域计算成本、降低密集应用开销、
兼顾云服务买卖双方利益并保证双层负载均衡性,也是基础设施运维商增强竞争优势所在.随着新常态(云计
算、异构计算)、新应用(计算密集型应用、数据密集型应用)和 QoS 新指标(能效)的出现,绿色感知的异构实时
调度研究具有重大的理论和应用价值.
针对现有元启发式算法为基础的实时调度算法大多存在的进化动力不足、个体多样性不够或收敛速度过
慢等缺陷,本文着眼于软硬件节能原理的深度融合,提出一种新的绿色异构调度算法.
大量仿真实验结果显示:无论对于数据密集型还是计算密集型实例,GHSA_di/II 算法较其他 3 种元启发式
异构调度算法,在整体性能、节能降耗以及可扩展性等方面,都具明显优势.
概言之,本研究是文献[11]的后续工作,而无论辐射广度、攻关难度、理论深度还是创新高度,本文的 GHSA_
di/II 算法都是其进一步的拓展和延伸,并可为推进云模式中软、硬件节能联动以提高其衔接实效探索新的道路,
或奠定一定的理论和技术基础.

References:
[1] Belkhir L, Elmeligi A. Assessing ICT global emissions footprint: Trends to 2040 & recommendations. Journal of Cleaner
Production, 2018,177:448−463.
[2] Ganji M, Kazemipoor H, Molana SMH, et al. A green multi-objective integrated scheduling of production and distribution with
heterogeneous fleet vehicle routing and time windows. Journal of Cleaner Production, 2020,259:120824.
[3] Constantinescu DA, Navarro A, Corbera F, et al. Efciency and productivity for decision making on low-power heterogeneous
CPU+GPU SoCs. The Journal of Supercomputing, https://doi.org/10.1007/s11227-020-03257-3
[4] Nejat M, Manivannan M, Pericàs M, et al. Coordinated management of DVFS and cache partitioning under QoS constraints to save
energy in multi-core systems. Journal of Parallel and Distributed Computing, 2020,144:246−259.
[5] Asghari A, Sohrabi MK, Yaghmaee F. Task scheduling, resource provisioning, and load balancing on scientifc workfows using
parallel SARSA reinforcement learning agents and genetic algorithm. The Journal of Supercomputing, https://doi.org/10.1007/
s11227-020-03364-1
[6] Saroja S, Revathi T. Hybrid dual-objective parallel genetic algorithm for heterogeneous multiprocessor scheduling. Cluster
Computing—The Journal of Networks Software Tools and Applications, 2020,23:441−450. https://doi.org/10.1007/s10586-019-
02934-0