[1]任明,沈达.基于深度学习的云平台动态自适应任务调度[J].计算机技术与发展,2024,34(08):17-22.[doi:10.20165/j.cnki.ISSN1673-629X.2024.0135]
　REN Ming,SHEN Da.Dynamic Adaptive Task Scheduling for Cloud Computing Platform Based on Deep Learning[J].,2024,34(08):17-22.[doi:10.20165/j.cnki.ISSN1673-629X.2024.0135]
点击复制

基于深度学习的云平台动态自适应任务调度

《计算机技术与发展》[ISSN:1006-6977/CN:61-1281/TN]

卷:
34
期数:
2024年08期
页码:
17-22
栏目:
大数据与云计算
出版日期:
2024-08-10

文章信息/Info

Title:
Dynamic Adaptive Task Scheduling for Cloud Computing Platform Based on Deep Learning
文章编号:
1673-629X(2024)08-0017-06
作者:
任明; 沈达
中国银联股份有限公司,上海 201201
Author(s):
REN Ming; SHEN Da
China UnionPay Co. ,Ltd. ,Shanghai 201201,China
关键词:
云计算; 深度学习; 任务调度; 自适应策略; 多头注意力; 模型选择
Keywords:
cloud computing; deep learning; task scheduling; adaptive strategy; multi-head attention; model selection
分类号:
TP301
DOI:
10.20165/j.cnki.ISSN1673-629X.2024.0135
摘要:
云计算环境下任务调度是优化云应用服务质量的热点研究问题,目前工业界和学术界重点关注任务调度策略。 然而,现有方法依赖运维人员的系统实现知识或复杂的深度神经网络,需要较高计算资源,产生更高执行成本,难以适应动态变化的多样化任务类型。 针对该问题,提出一种基于深度学习的云计算平台动态自适应任务调度策略。 首先,从待处理任务、可用云资源及系统运行状态等三方面提取任务调度特征;其次,构建深度学习模型对特征编码,通过多头图注意力机制推理解码以预测策略的任务处理和调度执行成本;最后,根据调度收益从策略集中选择当前最优任务调度策略,同时基于迭代反馈机制计算损失函数以在线优化模型。 建立虚拟化云计算服务器集群,实现典型的多种任务调度策略,模拟真实 AI 任务工作负载。 实验结果表明,所提出策略与现有实验选取方法相比能够有效降低响应时间、执行成本及运 行能耗。
Abstract:
Task scheduling in cloud computing environment is a hot research issue to optimize the quality of service of cloud applications.At present,industry and academia have proposed a variety of task scheduling strategies. Existing data-driven scheduling strategies rely on complex deep neural networks,which require high computing resources and incur higher execution costs,and are difficult to adapt to dy-namically changing diverse task types. To solve this problem, we propose a dynamic adaptive task scheduling strategy for cloud computing platform based on deep learning. Firstly,the task scheduling features were extracted from three aspects of pending tasks,available cloud resources and system running status. Then, a deep learning model was constructed to encode the features, and the execution cost and response time of the strategy were predicted through inference and decoding. Finally, the current optimal task scheduling strategy was selected from the strategy set according to the scheduling profit, and the online optimization model of loss function was calculated. The experimental results show that the proposed strategy improves the execution cost,response time and energy consumption compared with the existing methods.

相似文献/References:

[1]王茜,朱志祥,史晨昱,等.应用于数据库安全保护的加解密引擎系统[J].计算机技术与发展,2014,24(01):143.
　WANG Qian[],ZHU Zhi-xiang[],SHI Chen-yu[],et al.Encryption and Decryption Engine System Applying to Database Security and Detection[J].,2014,24(08):143.
[2]陈丹伟 黄秀丽 任勋益.云计算及安全分析[J].计算机技术与发展,2010,(02):99.
　CHEN Dan-wei,HUANG Xiu-li,REN Xun-yi.Analysis of Cloud Computing and Cloud Security[J].,2010,(08):99.
[3]孙放 陈云芳 林杭锋.适用于富客户端的云计算模型[J].计算机技术与发展,2010,(08):96.
　SUN Fang,CHEN Yun-fang,LIN Hang-feng.Cloud Computing Model Applicable to Rich Client Applications[J].,2010,(08):96.
[4]郭苑 张顺颐 孙雁飞.物联网关键技术及有待解决的问题研究[J].计算机技术与发展,2010,(11):180.
　GUO Yuan,ZHANG Shun-yi,SUN Yan-fei.Research of Key Technologies and Unresolved Questions of Internet of Things[J].,2010,(08):180.
[5]李玲娟 张敏.云计算环境下关联规则挖掘算法的研究[J].计算机技术与发展,2011,(02):43.
　LI Ling-juan,ZHANG Min.Research on Algorithms of Mining Association Rule under Cloud Computing Environment[J].,2011,(08):43.
[6]王德政 申山宏 周宁宁.云计算环境下的数据存储[J].计算机技术与发展,2011,(04):81.
　WANG De-zheng,SHEN Shan-hong,ZHOU Ning-ning.Data Storage in Cloud Computing Environment[J].,2011,(08):81.
[7]宋丽华 姜家轩 张建成 田长录 马文征.黄河三角洲云计算平台关键技术的研究[J].计算机技术与发展,2011,(06):40.
　SONG Li-hua,JIANG Jia-xuan,ZHANG Jian-cheng,et al.Research of Key Technologies of Cloud Computing of Yellow River Delta[J].,2011,(08):40.
[8]田宏伟 解福 倪俊敏.云计算环境下基于粒子群算法的资源分配策略[J].计算机技术与发展,2011,(12):22.
　TIAN Hong-wei,XIE Fu,NI Jun-min.Resource Allocation Algorithm Based on Particle Swarm Algorithm in Cloud Computing Environment[J].,2011,(08):22.
[9]张慧 邢培振.云计算环境下信息安全分析[J].计算机技术与发展,2011,(12):164.
　ZHANG Hui,XING Pei-zhen.Information Security Analysis in Cloud Computing Environment[J].,2011,(08):164.
[10]张建成[] 宋丽华[] 鹿全礼[] 郭锐[] 刘永泉[].云计算方案分析研究[J].计算机技术与发展,2012,(01):165.
　ZHANG Jian-cheng,SONG Li-hua,LU Quan-li,et al.Study and Analysis of Cloud Computing Procedure[J].,2012,(08):165.
[11]蒿敬波,阳广贤,肖湘江,等.基于 Transformer 模型的心音小波谱图识别[J].计算机技术与发展,2023,33(10):189.[doi:10. 3969 / j. issn. 1673-629X. 2023. 10. 029]
　HAO Jing-bo,YANG Guang-xian,XIAO Xiang-jiang,et al.Recognition of Heart Sound Scalograms Based on Transformer Model[J].,2023,33(08):189.[doi:10. 3969 / j. issn. 1673-629X. 2023. 10. 029]

更新日期/Last Update: 2024-08-10