Continuous performance optimization on odern Architectures

现代架构的持续性能优化

• 批准号: 536639-2018
• 负责人: Mashtizadeh, Ali
• 金额: $ 8.84万
• 依托单位: University of Waterloo
• 依托单位国家: 加拿大
• 项目类别: Collaborative Research and Development Grants
• 财政年份: 2020
• 资助国家: 加拿大
• 起止时间: 2020-01-01 至 2021-12-31
• 项目状态: 已结题
• 来源: https://www.nserc-crsng.gc.ca/ase-oro/Details-Detailles_eng.asp?id=699285
• 关键词: Continuous; performance; optimization; odern; Architectures

Today, multicore scalability poses increasing challenges to developers across all platforms from mobile to the data center. The intensive competition between the major processor vendors including Intel, AMD, IBM, and ARM is leading new innovation in computer architecture that systems software and application developers have only begun to take advantage of. There is an increased complexity of developing software on these platforms where developers are now exposed more than ever to the performance idiosyncrasies of the processor architecture. Impacting performance, performance isolation, and even security of their applications. The goal of this research is to provide the tools and insights to help developers make the best use of the hardware. We focus on three pieces: First, design and implement a continuous profiling system that will help developers make sense of the first, second and third order performance metrics available in modern processors. Second, build a portable runtime to help developers make optimal use of the new instructions that allow applications to control cache behavior, arguably one of the most complex issues with the largest gain for developers. Third, explore the use of this runtime and several new processor features for partitioning applications for performance, correctness, and security.

如今，多核可扩展性对跨所有平台的开发人员提出了越来越多的挑战， 移动的到数据中心。主要处理器供应商之间的激烈竞争， 英特尔、AMD、IBM和ARM正在引领计算机架构的新创新， 软件和应用程序开发人员才刚刚开始利用。存在增加 在这些平台上开发软件的复杂性， 处理器架构的性能特性。影响性能， 性能隔离，甚至应用程序的安全性。 这项研究的目标是提供工具和见解，以帮助开发人员做出最好的 硬件的使用。我们重点做了三件事：第一，设计和实现一个连续的剖析 一个帮助开发人员理解一阶、二阶和三阶性能指标的系统 在现代处理器中。第二，构建可移植的运行时，帮助开发人员实现最佳的 使用允许应用程序控制缓存行为的新指令，可以说是 最复杂的问题，为开发人员带来最大的收益。第三，探索此运行时的使用 以及几个新的处理器特性，用于根据性能、正确性和 安全