StreamSys: A Lightweight Executable Delivery System for Edge Computing

IF 5.3 2区计算机科学 Q1 COMPUTER SCIENCE, INFORMATION SYSTEMS

IEEE Transactions on Cloud Computing Pub Date : 2024-12-24 DOI:10.1109/TCC.2024.3521978

Jun Lu;Zhenya Ma;Yinggang Gao;Sheng Yue;Ju Ren;Yaoxue Zhang

{"title":"StreamSys: A Lightweight Executable Delivery System for Edge Computing","authors":"Jun Lu;Zhenya Ma;Yinggang Gao;Sheng Yue;Ju Ren;Yaoxue Zhang","doi":"10.1109/TCC.2024.3521978","DOIUrl":null,"url":null,"abstract":"Edge computing brings several challenges when it comes to data movement. First, moving large data from edge devices to the server is likely to waste bandwidth. Second, complex data patterns (e.g., traffic cameras) on devices require flexible handling. An ideal approach is to move code to data instead. However, since only a small portion of code is required, moving the executable as well as their libraries to the devices can be an overkill. While loading code on demand from remote such as NFS can be a stopgap, but on the other hand leads to low efficiency for irregular access patterns. This article presents <sc>StreamSys</small>, a lightweight executable delivery system that loads code on demand by redirecting the local disk IO to the server through optimized network IO. We employ a Markov-based prefetch mechanism on the server side. It learns the access pattern of code and predicts the block sequence for the client to reduce the network round trip. Meanwhile, server-side <sc>StreamSys</small> asynchronously prereads the block sequence from the disk to conceal disk IO latency beforehand. Evaluation shows that the latency of <sc>StreamSys</small> is up to 71.4% lower than the native Linux file system based on SD card and up to 62% lower than NFS in wired environments.","PeriodicalId":13202,"journal":{"name":"IEEE Transactions on Cloud Computing","volume":"13 1","pages":"213-226"},"PeriodicalIF":5.3000,"publicationDate":"2024-12-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"IEEE Transactions on Cloud Computing","FirstCategoryId":"94","ListUrlMain":"https://ieeexplore.ieee.org/document/10814051/","RegionNum":2,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"COMPUTER SCIENCE, INFORMATION SYSTEMS","Score":null,"Total":0}

引用次数: 0

Abstract

Edge computing brings several challenges when it comes to data movement. First, moving large data from edge devices to the server is likely to waste bandwidth. Second, complex data patterns (e.g., traffic cameras) on devices require flexible handling. An ideal approach is to move code to data instead. However, since only a small portion of code is required, moving the executable as well as their libraries to the devices can be an overkill. While loading code on demand from remote such as NFS can be a stopgap, but on the other hand leads to low efficiency for irregular access patterns. This article presents StreamSys, a lightweight executable delivery system that loads code on demand by redirecting the local disk IO to the server through optimized network IO. We employ a Markov-based prefetch mechanism on the server side. It learns the access pattern of code and predicts the block sequence for the client to reduce the network round trip. Meanwhile, server-side StreamSys asynchronously prereads the block sequence from the disk to conceal disk IO latency beforehand. Evaluation shows that the latency of StreamSys is up to 71.4% lower than the native Linux file system based on SD card and up to 62% lower than NFS in wired environments.

查看原文本刊更多论文

StreamSys：用于边缘计算的轻量级可执行文件交付系统

当涉及到数据移动时，边缘计算带来了几个挑战。首先，将大数据从边缘设备移动到服务器可能会浪费带宽。其次，设备上复杂的数据模式（如交通摄像头）需要灵活处理。一种理想的方法是将代码移到数据中。然而，由于只需要一小部分代码，因此将可执行文件及其库移动到设备上可能是多余的。虽然从远程（如NFS）按需加载代码可能是权宜之计，但另一方面，对于不规则的访问模式会导致效率低下。本文介绍了StreamSys，这是一个轻量级的可执行交付系统，它通过优化的网络IO将本地磁盘IO重定向到服务器，按需加载代码。我们在服务器端采用了基于markov的预取机制。它学习代码的访问模式，并为客户端预测块序列，以减少网络往返。同时，服务器端StreamSys异步地从磁盘预读块序列，以预先隐藏磁盘IO延迟。评估表明，StreamSys的延迟比基于SD卡的本地Linux文件系统低71.4%，比有线环境下的NFS低62%。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

IEEE Transactions on Cloud Computing Computer Science-Software

CiteScore

9.40

自引率

6.20%

发文量

167

期刊介绍： The IEEE Transactions on Cloud Computing (TCC) is dedicated to the multidisciplinary field of cloud computing. It is committed to the publication of articles that present innovative research ideas, application results, and case studies in cloud computing, focusing on key technical issues related to theory, algorithms, systems, applications, and performance.