PPoSS: Planning: Prescriptive Memory: Razing the Semantic Wall Between Applications and Computer Systems

PPoSS：规划：规定性记忆：消除应用程序和计算机系统之间的语义墙

• 批准号: 2028949
• 负责人: Phillip Gibbons
• 金额: $ 25万
• 依托单位: Carnegie-Mellon University
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2020
• 资助国家: 美国
• 起止时间: 2020-10-01 至 2022-09-30
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=2028949&HistoricalAwards=false
• 关键词: PPoSS; Planning; Prescriptive; Memory; Razing

There is a large semantic gap between the high-level understanding of an application's goals and the sequence of processor instructions and memory accesses observed by computer hardware. This is due to the long-standing narrow interface between software and the computing platform it is running on, which acts as a semantic wall that prevents the platform from meeting application needs. The cost of this wall is growing with the rapid increase in specialized hardware accelerators and new classes of memory and storage. This project razes the semantic wall by developing a fundamentally more expressive interface, called prescriptive memory, which enables more semantic information to be conveyed from applications to system software to hardware. Prescriptive memory will enable applications to take advantage of new hardware accelerators and new memory technologies, spurring market acceptance, which in turn spurs industry investment both in more advanced accelerators and memories and in application software that uses them. This virtuous cycle boosts the entire high tech ecosystem, which in turn boosts all aspects of society that benefit from improved computer performance. Prescriptive memory will also improve the energy-efficiency of computations in data centers, reducing their carbon footprint. As a general and powerful abstraction, prescriptive memory will provide significant--often game changing--benefits to a wide range of applications. The project will demonstrate such benefits for important applications in robotics and machine learning, with a showcase application being training robots to perform complicated tasks with and among humans in cluttered environments, safely and effectively.The project will define a prescriptive memory abstraction that enables higher-level properties to be associated with dynamic regions of a program's memory. It is through these properties that semantic intent, application goals, application preferences, etc. are expressed and conveyed throughout the software stack down to the hardware. The project will design and implement an open source software stack that supports and benefits from the new abstraction. A unifying hardware mechanism will be developed that enables hardware to leverage these properties to reduce overheads and better meet application needs. A super-controller will be designed and implemented that, based on these properties and dynamic system state information, makes (near-)optimal decisions on data placement and accelerator use at every point in time. The project will explore how prescriptive memory benefits a range of applications bottlenecked by the semantic wall, with a specific focus on providing orders of magnitude improvements in the speed and scale (fidelity) of robotic worlds simulations, a key to enabling safe and effective "AI in the wild."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.

在对应用程序目标的高级理解与计算机硬件观察到的处理器指令和存储器访问序列之间存在很大的语义差距。这是由于软件和它运行的计算平台之间长期存在的狭窄接口，这就像一堵语义墙，阻止了平台满足应用程序的需求。随着专用硬件加速器和新型内存和存储的快速增长，这堵墙的成本也在增长。这个项目通过开发一个从根本上更具表现力的接口（称为规定性内存）来雷兹语义墙，该接口使更多的语义信息能够从应用程序传递到系统软件再到硬件。 规定内存将使应用程序能够利用新的硬件加速器和新的内存技术，从而刺激市场接受，这反过来又刺激了行业对更先进的加速器和内存以及使用它们的应用软件的投资。 这种良性循环促进了整个高科技生态系统的发展，反过来又促进了社会的各个方面，这些方面都受益于计算机性能的提高。 规定内存还将提高数据中心计算的能源效率，减少碳足迹。 作为一种通用且强大的抽象，规定性内存将为广泛的应用程序提供显着的（通常是改变游戏规则的）好处。 该项目将展示机器人和机器学习的重要应用的好处，展示应用是训练机器人在混乱的环境中安全有效地与人类一起执行复杂的任务。该项目将定义一个规定性的内存抽象，使更高级别的属性与程序内存的动态区域相关联。正是通过这些属性，语义意图、应用目标、应用偏好等在整个软件堆栈中被表达和传达到硬件。 该项目将设计和实现一个开源软件栈，支持并受益于新的抽象。 将开发一种统一的硬件机制，使硬件能够利用这些属性来减少开销并更好地满足应用需求。 将设计和实现一个超级控制器，根据这些属性和动态系统状态信息，在每个时间点对数据放置和加速器使用做出（接近）最佳决策。 该项目将探索指令性记忆如何使一系列受到语义墙检查的应用程序受益，特别关注在机器人世界模拟的速度和规模（保真度）方面提供数量级的改进，这是实现安全有效的“人工智能”的关键。“这个奖项反映了NSF的法定使命，并已被认为是值得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的支持。

项目成果

Brief Announcement: Block-Granularity-Aware Caching

简短公告：块粒度感知缓存

• DOI: 10.1145/3409964.3461818
• 发表时间: 2021
• 期刊: SPAA '21: 33rd ACM Symposium on Parallelism in Algorithms and Architectures
• 作者: Beckmann, Nathan;Gibbons, Phillip B.;McGuffey, Charles
• 通讯作者: McGuffey, Charles

Cortex: A Compiler for Recursive Deep Learning Models

• 发表时间: 2020-11
• 期刊: ArXiv
• 作者: Pratik Fegade;Tianqi Chen;Phillip B. Gibbons;T. Mowry

Brief Announcement: Spatial Locality and Granularity Change in Caching

简短公告：缓存的空间局部性和粒度变化

• DOI: 10.1145/3490148.3538559
• 发表时间: 2022
• 期刊: SPAA '22: 34th ACM Symposium on Parallelism in Algorithms and Architectures
• 作者: Beckmann, Nathan;Gibbons, Phillip B.;McGuffey, Charles
• 通讯作者: McGuffey, Charles

RTRBench: A Benchmark Suite for Real-Time Robotics

• DOI: 10.1109/ispass55109.2022.00024
• 发表时间: 2022-05
• 期刊: 2022 IEEE International Symposium on Performance Analysis of Systems and Software (ISPASS)
• 作者: Mohammad Bakhshalipour;M. Likhachev;Phillip B. Gibbons

The Processing-in-Memory Model

内存处理模型

• DOI: 10.1145/3409964.3461816
• 发表时间: 2021
• 期刊: SPAA '21: 33rd ACM Symposium on Parallelism in Algorithms and Architectures
• 作者: Kang, Hongbo;Gibbons, Phillip B.;Blelloch, Guy E.;Dhulipala, Laxman;Gu, Yan;McGuffey, Charles
• 通讯作者: McGuffey, Charles