CNS Core: Small: Scaling Coordination for Microservices with Pangaea

CNS 核心：小型：Pangea 微服务的扩展协调

• 批准号: 1910390
• 负责人: Wyatt Lloyd
• 金额: $ 50万
• 依托单位: Princeton University
• 依托单位国家: 美国
• 项目类别: Continuing Grant
• 财政年份: 2019
• 资助国家: 美国
• 起止时间: 2019-10-01 至 2023-09-30
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1910390&HistoricalAwards=false
• 关键词: CNS; Core; Small; Scaling; Coordination

An increasingly popular way to design applications is using an architecture that splits out the functionality of an application into many distinct micro-services. It remains challenging, however, to implement micro-services that store and update because building them requires solving many hard problems each time: providing fault tolerance, handling many concurrent operations, and scaling up the work they can do. This proposal aims to solve these hard problems in a way that can be reused across many micro-services that store and update state. It builds on an emerging design that uses a single machine, called a sequencer, to order operations to state. Compared to more traditional designs, sequencers provide better scaling of the work the micro-service does, but make providing fault tolerance and handling concurrency harder. This proposal seeks to address the shortcomings of sequencers by finding a general way to provide fault tolerance, tame concurrency, and provide even greater scaling.The proposed work has the potential to improve the performance and decrease the development time of micro-service-based systems. Given the popularity of micro-service-based systems, this has the potential to improve many applications that are an integral part of modern life. The project plans include developing software prototypes and working with large Internet services to increase the likelihood of technology transfer. In addition, the research in the project will be conducted by graduate students and undergraduate students that will learn how micro-service-based systems work as well as how to build and improve them. Each of these skills is likely to be increasingly important in the future.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

设计应用程序的一种越来越流行的方法是使用一种架构，将应用程序的功能拆分为许多不同的微服务。然而，实现存储和更新的微服务仍然具有挑战性，因为构建它们每次都需要解决许多难题：提供容错，处理许多并发操作，以及扩展它们可以做的工作。该提案旨在以一种可以在许多存储和更新状态的微服务中重用的方式解决这些难题。它建立在一种新兴的设计之上，这种设计使用一台称为定序器的机器来对状态的操作进行排序。与更传统的设计相比，定序器提供了微服务所做工作的更好扩展，但提供容错和处理并发更困难。该提案旨在通过找到一种通用的方法来提供容错，驯服并发，并提供更大的scaling.The拟议的工作有可能提高性能和减少基于微服务的系统的开发时间，以解决定序器的缺点。鉴于基于微服务的系统的流行，这有可能改善许多作为现代生活不可或缺的一部分的应用程序。项目计划包括开发软件原型和与大型互联网服务合作，以增加技术转让的可能性。此外，该项目的研究将由研究生和本科生进行，他们将学习基于微服务的系统如何工作以及如何构建和改进它们。这一奖项反映了NSF的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

项目成果

Facebook's Tectonic Filesystem: Efficiency from Exascale

• 发表时间: 2021
• 作者: Satadru Pan;Theano Stavrinos;Yunqiao Zhang;Atul Sikaria;Pavel Zakharov;Abhinav Sharma;Shiva Shankar;Michael Shuey;R. Wareing;Monika Gangapuram;Guanglei Cao;Christian Preseau;Pratap Singh;K. Patiejunas;JR Tipton;Ethan Katz-Bassett;Wyatt Lloyd

[Solution] Mason: Scalable, Contiguous Sequencing for Building Consistent Services

[解决方案] Mason：可扩展、连续的排序以构建一致的服务

• DOI: 10.5070/sr33161354
• 发表时间: 2024
• 期刊: Journal of Systems Research
• 作者: Hodsdon, Christopher;Stavrinos, Theano;Katz-Bassett, Ethan;Lloyd, Wyatt
• 通讯作者: Lloyd, Wyatt