Utilizing Memory Parallelism for High Performance Data Processing

虽然微处理器设计的进步继续提高计算速度，但计算系统的数据访问速度的改进远远落后。与此同时，信息检索、计算机动画、大数据分析等数据密集型大规模应用正在兴起。数据访问延迟已成为现代高性能计算（HPC）的重要性能瓶颈。内存并发存在于现代内存层次结构的每一层；然而，传统的计算系统主要是为了提高CPU利用率而设计的，并且在解决HPC中数据移动的关键问题方面具有固有的局限性。在这项研究中，PI提出了内存并行的概念和定制并行内存（CuPM）系统架构，以解决数据移动瓶颈问题。CuPM是一种架构，旨在利用内存并发性来支持高性能数据处理（HPDP）。这个新的抽象通过构建新的Concurrent-AMAT度量、一组新的数据访问并发性理论结果、一系列最近在并行I/O优化方面取得的成功以及新的技术机会，提取了并行内存系统的一般原则。CuPM是一种新颖的跨层和横切方法，用于基于内存并发性带来的机会来探索和利用当前和未来的内存技术。当前的HPC系统不能完全识别或利用现代内存系统提供的并发性。CuPM的关键新思想是建立一种系统的方法来探索、增强、利用和定制内存并发，从而构建有效的内存系统。它的开发有两个目标：理解和揭示内存并发性及其特性，探索和利用当前的内存并发性。前者将评价后者的潜力；而后者将验证和测试前者。总之，它们使CuPM能够解决极限计算的核心挑战。高效的存储系统。所建议的CuPM体系结构具有从应用程序到运行时再到体系结构转换系统的潜力。

While advances in microprocessor design continue to increase computing speed, improvements in data access speed of computing systems lag far behind. At the same time, data-intensive large-scale applications, such as information retrieval, computer animation, and big data analytics are emerging. Data access delay has become the vital performance bottleneck of modern high performance computing (HPC). Memory concurrency exists at each layer of modern memory hierarchies; however, conventional computing systems are primarily designed to improve CPU utilization and have inherent limitations in addressing the critical issue of data movement in HPC. In this research, the PI proposes the concept of memory parallelism and the customized parallel memory (CuPM) system architecture in order to address data movement bottleneck issues. CuPM is an architecture designed to exploit memory concurrency to support high performance data processing (HPDP). This new abstraction extracts the general principles of parallel memory system by building on the new Concurrent-AMAT metric, a set of fresh theoretical results in data access concurrency, a series of recent successes in parallel I/O optimization, and new technology opportunities. CuPM is a novel cross-layer and cross-cutting approach for exploring and utilizing current and future memory technologies based on opportunities arising from memory concurrency.Current HPC systems do not fully recognize or exploit the concurrency that modern memory systems provide. The key new idea of CuPM is to establish a systematic way of exploring, enhancing, utilizing, and customizing memory concurrency to build effective memory systems. It will be developed with two goals: to understand and reveal the memory concurrency and its properties, and to explore and utilize current memory concurrency for HPDP. The former will evaluate the potential of the latter; while the latter will verify and testing the former. Together, they enable CuPM to address a central challenge for extreme computing ? efficient memory systems. The proposed CuPM architecture has the potential to transform systems from application through runtime through architecture.