CSR: Small: Protocols and Languages for Low-Coordination Consistent Geodistributed Transactions

CSR：小型：低协调一致地理分布式事务的协议和语言

• 批准号: 1717554
• 负责人: Andrew Myers
• 金额: $ 48.68万
• 依托单位: Cornell University
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2017
• 资助国家: 美国
• 起止时间: 2017-10-01 至 2022-09-30
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1717554&HistoricalAwards=false
• 关键词: CSR; Small; Protocols; Languages; Low

The world now runs on applications that operate at global scale, available everywhere. The information that these applications use must also be distributed around the globe so it is close to application users. However, keeping this geo-distributed information in sync is a major challenge for application developers. They must choose between building software in a way that is slow but correct or fast but error-prone. This project aims to make it easier for developers to build reliable, high-performance distributed applications. Developers will be able to write code in a manner similar to the way they would when using a database that offers simple reasoning and strong guarantees, but the project will introduce new software architectures and tools that improve performance. The project will demonstrate using real software that its innovations are effective in practice.This project aims for broad impact on the reliability and performance of distributed software. Its success will enable a larger class of developers to build distributed software systems with rich functionality and tight integration. Currently such systems require expert engineers and can be undertaken only by a few large software companies. Democratizing construction of distributed systems will empower a wide range of developers at government agencies, health care and nonprofit organizations, and medium-sized businesses. The project will release software artifacts embodying the technical advances and will demonstrate their use to build real software. Developed software will be publicly available at the project home page, www.cs.cornell.edu/projects/proteus, useful not only to researchers but also to developers.

世界现在运行在全球范围内运行的应用程序上，无处不在。这些应用程序所使用的信息也必须分布在地球仪的各个角落，以便更接近应用程序用户。 然而，保持这种地理分布的信息同步是应用程序开发人员面临的主要挑战。他们必须在缓慢但正确的方式和快速但容易出错的方式之间做出选择。 该项目旨在使开发人员更容易构建可靠，高性能的分布式应用程序。开发人员将能够以类似于使用提供简单推理和强有力保证的数据库的方式编写代码，但该项目将引入新的软件架构和工具来提高性能。该项目将使用真实的软件证明其创新在实践中是有效的。该项目旨在对分布式软件的可靠性和性能产生广泛的影响。 它的成功将使更多的开发人员能够构建具有丰富功能和紧密集成的分布式软件系统。目前，这种系统需要专家工程师，只能由少数大型软件公司承担。分布式系统的民主化构建将使政府机构、医疗保健和非营利组织以及中型企业的广泛开发人员获得授权。该项目将发布体现技术进步的软件工件，并将展示它们用于构建真实的软件。 开发的软件将在项目主页www.cs.cornell.edu/projects/proteus上公开提供，不仅对研究人员有用，对开发人员也有用。

项目成果

PDL: a high-level hardware design language for pipelined processors

PDL：用于流水线处理器的高级硬件设计语言

• DOI: 10.1145/3519939.3523455
• 发表时间: 2022
• 期刊: ACM SIGPLAN Conf. on Programming Language Design and Implementation (PLDI
• 作者: Zagieboylo, Drew;Sherk, Charles;Suh, Gookwon Edward;Myers, Andrew C.
• 通讯作者: Myers, Andrew C.

MixT: a language for mixing consistency in geodistributed transactions

• DOI: 10.1145/3192366.3192375
• 发表时间: 2018-06
• 期刊: Proceedings of the 39th ACM SIGPLAN Conference on Programming Language Design and Implementation
• 作者: Mae Milano;A. Myers

A flexible type system for fearless concurrency

灵活的类型系统，可实现无所畏惧的并发

• DOI: 10.1145/3519939.3523443
• 发表时间: 2022
• 期刊: ACM SIGPLAN Conf. on Programming Language Design and Implementation (PLDI
• 作者: Milano, Mae;Turcotti, Joshua;Myers, Andrew C.
• 通讯作者: Myers, Andrew C.

A tour of Gallifrey, a language for geodistributed programming

Gallifrey 之旅，一种地理分布式编程语言

• 发表时间: 2019
• 期刊: Summit on Advances in Programming Languages
• 作者: Milano, Matthew;Recto, Rolph;Magrino, Tom;Myers, Andrew C.
• 通讯作者: Myers, Andrew C.

Efficient, Consistent Distributed Computation with Predictive Treaties

具有预测性条约的高效、一致的分布式计算

• 发表时间: 2019
• 期刊: EuroSys 2019
• 作者: Magrino, Tom;Liu, Jed;Foster, Nate;Gehrke, Johannes;Myers, Andrew C.
• 通讯作者: Myers, Andrew C.