视频监控图像资源调度算法研究与设计  

Research and Design of a Video Image Resource Scheduling Algorithm

在线阅读下载全文

作  者:吴志雄 崔勇[1] WU Zhixiong;CUI Yong(Department of Computer Science and Technology,Tsinghua University,Beijing 100084,China)

机构地区:[1]清华大学计算机科学与技术系,北京100084

出  处:《小型微型计算机系统》2024年第12期3035-3041,共7页Journal of Chinese Computer Systems

基  金:2022年福建省工业和信息化厅软件业技术创新重点攻关及产业化项目(闽工信函软件[2022]216号)资助。

摘  要:针对公共安全系统中视频图像解析过程中存在的资源利用率低和系统吞吐量不足的问题,本文提出了一种基于帕累托最优前沿的资源调度算法.该算法通过对问题定义分析,构建了一个多目标资源调度模型,该模型根据Pod的需求与可用资源之间的关系,建立两个优化目标:最小化工作节点的资源损失率,最大化Pod之间的亲和性,从而提高Pod的系统吞吐量和响应时间.在此基础上,本文设计了对多目标进行乘积作为求解目标函数,采用帕累托最优前沿免疫算法对调度模型进行求解,实现资源在各Pod之间的最优分配,从而满足业务需求并最小化资源使用.实验结果表明,与传统方法相比,本文提出的算法在适应度指标上具有一定优势,能够有效提升资源利用率和Pod间的整体运行性能,最终实现视频图像解析的效率和性能的提升.To address the issues of l low resource utilization and insufficient system throughput in video image parsing in public safety systems this paper proposes a resource scheduling algorithm based on Pareto optimal frontier.This algorithm constructs a multi-objective resource scheduling model by analyzing the problem definition.Based on the relationship between Pod demand and available resources,the model establishes two optimization objectives:minimizing the resource loss rate of work nodes and maximizing the affinity between Pods,thereby improving the system throughput and response time of Pods.On this basis,this article designs the multiplication of multiple objectives as the objective function for solving the scheduling model.The Pareto optimal frontier immune algorithm is used to solve the scheduling model,achieving the optimal allocation of resources among each Pod,thereby meeting business needs and minimizing resource usage.The experimental results show that compared with traditional methods,the algorithm proposed in this paper has certain advantages in fitness indicators,which can effectively improve resource utilization and overall performance among Pods,ultimately achieving an improvement in the efficiency and performance of video image parsing.

关 键 词:资源调度 免疫算法 资源亲和度 多目标问题 

分 类 号:TP393[自动化与计算机技术—计算机应用技术]

 

参考文献:

正在载入数据...

 

二级参考文献:

正在载入数据...

 

耦合文献:

正在载入数据...

 

引证文献:

正在载入数据...

 

二级引证文献:

正在载入数据...

 

同被引文献:

正在载入数据...

 

相关期刊文献:

正在载入数据...

相关的主题
相关的作者对象
相关的机构对象