面向流式数据的弹性缓存管理系统设计  

Adaptive management for backpressure in data stream caching

作　　者：陶雨棚 陈庆奎[1] 黄陈[2] TAO Yupeng;CHEN Qingkui;HUANG Chen(School of Optical-Electrical and Computer Engineering,University of Shanghai for Science and Technology,Shanghai 200093,China;Shanghai General Hospital,Shanghai 201620,China)

机构地区：[1]上海理工大学光电信息与计算机工程学院,上海200093 [2]上海市第一人民医院,上海201620

出　　处：《智能计算机与应用》2024年第2期62-68,共7页Intelligent Computer and Applications

基　　金：国家自然科学基金(61572325);上海重点科技攻关项目(19DZ1208903);上海交通大学医学院医工交叉项目。

摘　　要：腹腔手术过程中需要维持腹腔的恒压,在利用物联网技术和云计算精准、实时地辅助医生完成恒压的过程中,存在大量流式数据,一些高负荷场景中存在缓存数据积压问题,导致的系统延迟将影响医疗器械的稳定性。在腹腔镜手术恒压器械中,缓存发挥着至关重要的作用,可以提高系统的吞吐量和响应速度,并减少对下游系统的压力。在实际应用中,上游传感器端由于网络延迟、数据内容等因素导致该单元的数据产出往往是不稳定的,在这种场景下常用的静态缓存策略易造成数据积压或缓存浪费,进而影响资源利用率。为此,本文提出以协调集群中上、下游节点间的资源为基础,通过数据积压预测结合协同均衡的策略,下游单元能感知上游单元数据产出后的负载变化趋势,预先调整资源分配;上游能通过管理单元的均衡调整,及时调整目标缓存节点及数据处理率。利用弹性缓存管理后,缓存占用比不随上游传感器数据量的变化而变化,数据得到了及时处理。In the process of abdominal surgery,the maintenance of constant pressure in the abdominal cavity requires a lot of sensor information to be accurately and real-time assisted by the doctor using IoT and cloud computing technology.In some high-load scenarios,there is a problem of cache data accumulation,which leads to system delays and affects the stability of medical devices.In this abdominal laparoscopic constant pressure device,the cache plays a crucial role.It can improve the system′s throughput and response speed and reduce pressure on the downstream system.In actual use,the upstream sensor end often experiences unstable data output due to network delays,data content,and other factors.In this scenario,common static cache strategies are prone to data accumulation or cache waste,affecting resource utilization.Therefore,this paper proposes a strategy of data accumulation prediction combined with cooperative balance based on coordinating resources between upstream and downstream nodes in the cluster.In this management plan,the downstream unit can perceive the load change trend after the data output from the upstream unit and adjust resources beforehand,and the upstream unit can adjust the target cache node and data processing rate in a timely manner through the balance adjustment of the management unit.After using elastic cache management,the cache occupation ratio does not change with the upstream sensor data,and the data is processed in a timely manner.

关 键 词：缓存管理 弹性 数据积压问题 流式数据 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]