SPX: Collaborative Research: Harnessing the Power of High-Bandwidth Memory via Provably Efficient Parallel Algorithms

SPX：协作研究：通过可证明高效的并行算法利用高带宽内存的力量

基本信息

批准号：
1725543
负责人：
Michael Ferdman
金额：
$ 50万
依托单位：
SUNY at Stony Brook
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：
2017-09-15 至 2022-08-31
项目状态：
已结题

来源：
https://www.nsf.gov/awardsearch/showAward?AWD_ID=1725543&HistoricalAwards=false
关键词：
SPX Collaborative Research Harnessing Power

项目摘要

An important bottleneck for many parallel scientific applications is memory performance. Recently, vendors have introduced a new memory called high-bandwidth memory (HBM) as an approach to alleviate this bottleneck. This project will develop an algorithmic foundation for using HBM. The project has the potential for a broad economic, technological, and scientific impact, since industry has an investment in this technology and many of the nation's strategic codes are being run on HBM-capable machines. The PIs will integrate research with education at the graduate and undergraduate levels by training PhD, MS, and honors program BS students in cross-cutting issues encompassing algorithm design, high-performance software, and processor architecture. The new approach offered by vendors is to bond memory (HBM) directly to the processor chip, which allows for more connections, enabling higher bandwidth. Although the size of the new memory is larger than modern on-chip caches, physical constraints limit the capacity of the memory to be significantly smaller than DRAM. HBM does not cleanly fit in the standard memory hierarchy. This project will develop a foundational understanding of how to algorithmically design codes for HBM enhanced architectures. Overcoming these intellectual challenges to achieve multi-core scalability using HBM requires new algorithms, models, and abstractions, spearheaded by this collaboration between researchers who study hardware issues, high performance computing challenges, and theoretical modeling and analysis.

许多平行科学应用的重要瓶颈是记忆性能。最近，供应商推出了一种名为“高带记忆”（HBM）的新内存，以减轻这种瓶颈。该项目将为使用HBM建立算法基础。该项目具有广泛的经济，技术和科学影响的潜力，因为行业对这项技术进行了投资，并且该国许多战略代码都在具有HBM能力的机器上运行。 PI将通过培训博士学位，MS和荣誉计划BS学生将研究与研究生和本科级别的教育整合到跨记录问题中，包括算法设计，高性能软件和处理器体系结构。供应商提供的新方法是将内存（HBM）直接粘合到处理器芯片，这允许更多连接，从而实现更高的带宽。尽管新内存的大小大于现代芯片固定库，但物理限制限制了内存的能力明显小于DRAM。 HBM在标准内存层次结构中不干净地拟合。该项目将对如何为HBM增强体系结构设计算法设计代码有基本的理解。克服这些智力挑战以使用HBM实现多核可伸缩性，需要新的算法，模型和抽象，并以研究硬件问题，高性能计算挑战以及理论建模和分析的研究人员之间的合作为首。

项目成果

期刊论文数量（19）

专著数量（0）

科研奖励数量（0）

会议论文数量（0）

专利数量（0）

Timely Reporting of Heavy Hitters using External Memory

使用外部存储器及时报告重要人物

DOI：
发表时间：
2021
期刊：
SIGMOD record
影响因子：
1.1
作者：
Singh, S;Pandey, P;Bender, M.;Berry, J.;Farach-Colton, M;Johnson, R;Kroeger, T.;Phillips, C
通讯作者：
Phillips, C

Online Parallel Paging with Optimal Makespan

DOI：
10.1145/3490148.3538577
发表时间：
2022-07
期刊：
Proceedings of the 34th ACM Symposium on Parallelism in Algorithms and Architectures
影响因子：
0
作者：
Kunal Agrawal;M. A. Bender;Rathish Das;William Kuszmaul;E. Peserico;Michele Scquizzato
通讯作者：
Kunal Agrawal;M. A. Bender;Rathish Das;William Kuszmaul;E. Peserico;Michele Scquizzato

When Are Cache-Oblivious Algorithms Cache Adaptive? A Case Study of Matrix Multiplication and Sorting

缓存无关算法何时具有缓存自适应性？

DOI：
发表时间：
2022
期刊：
ESA
影响因子：
0
作者：
Bhattacharya, Arghya;Chowdhury, Abiyaz;Xu, Helen;Das, Rathish;Chowdhury, Rezaul Alam;Johnson, Rob;Nithyanand, Rishab;Bender, Michael A.
通讯作者：
Bender, Michael A.

Automatic HBM Management: Models and Algorithms

DOI：
10.1145/3490148.3538570
发表时间：
2022-07
期刊：
Proceedings of the 34th ACM Symposium on Parallelism in Algorithms and Architectures
影响因子：
0
作者：
Daniel DeLayo;Kenny Zhang;Kunal Agrawal;M. A. Bender;Jonathan W. Berry;Rathish Das;Benjamin Moseley;C. Phillips
通讯作者：
Daniel DeLayo;Kenny Zhang;Kunal Agrawal;M. A. Bender;Jonathan W. Berry;Rathish Das;Benjamin Moseley;C. Phillips

What Does Dynamic Optimality Mean in External Memory?

动态最优性在外部存储器中意味着什么？