深入理解IoTDB在时序数据工作负载下的性能表现  

In-depth Understanding on Performance of IoTDB in Handling Time-series Data Workloads

作　　者：刘健[1] 程建勋 闫高锋 吕泽涛 孙国道[1] 梁荣华[1] 蒋莉[1] LIU Jian;CHENG Jianxun;YAN Gaofeng;L Zetao;SUN Guodao;LIANG Ronghua;JIANG Li(College of Computer Science and Technology,Zhejiang University of Technology,Hangzhou 310023,China)

机构地区：[1]浙江工业大学计算机科学与技术学院,杭州310023

出　　处：《小型微型计算机系统》2025年第4期1014-1024,共11页Journal of Chinese Computer Systems

基　　金：国家自然科学基金项目(62202430)资助;浙江省自然科学基金杰出青年科学基金项目(LR23F020003)资助;浙江省基础公益研究计划项目(LTGG23F020005)资助.

摘　　要：IoT设备的大量部署导致时序数据呈现爆炸式增长,为了更好地管理大规模时序数据,高效满足上层时序用户的高性能访问需求,各种时序数据库被广泛开发和部署.但目前从业人员对于时序数据库的理解还不够深入,积累的知识和经验往往局限于传统数据库,并不适用于新兴的时序数据库.本文以面向IoT设计的开源时序数据库IoTDB为基础,利用基准测试工具IoT-Benchmark生成多样化的时序数据工作负载,深入探究IoTDB在处理时序数据工作负载时的性能表现和行为特征.本文的研究能够为时序数据库的设计者和开发者在开发和部署高性能时序数据库方面提供更有价值的帮助和指导.Time-series data is growing explosively driven by the rapid proliferation of IoT devices.To more efficiently manage the large-scale time-series data and meet the high-performance access demands of upper-layer time-series users,various time-series databases have been developed and deployed.Despite these efforts,practitioners currently lack a thorough understanding of time-series databases remains insufficient,with accumulated knowledge and experience predominantly confined to traditional databases.This limitation hinders the seamless adaptation of practices to emerging time-series databases.This paper focuses on addressing these challenges by leveraging the open-source time-series database,IoTDB,specifically designed for IoT applications.Employing the benchmarking tool IoT-Benchmark,diverse time-series data workloads can be generated for the comprehensive evaluation on the performance of IoTDB.The objective is to conduct a thorough investigation into the performance behaviors of IoTDB in handling time-series data workloads.The research in this paper can provide insightful guidance and help for system designers and developers to develop and deploy high-performance time-series databases.

关 键 词：IoTDB 时序数据工作负载 性能评估 新硬件 

分 类 号：TP392[自动化与计算机技术—计算机应用技术]