XPS: EXPL: Write Locality Theory and Optimization for Hybrid Memory

XPS：EXPL：混合内存的写入局部性理论和优化

• 批准号: 1629376
• 负责人: Chen Ding
• 金额: $ 30万
• 依托单位: University of Rochester
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2016
• 资助国家: 美国
• 起止时间: 2016-09-01 至 2019-08-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1629376&HistoricalAwards=false
• 关键词: XPS; EXPL; Write; Locality; Theory

A limit on extreme scale computing is the capacity, speed, durability and most importantly, energy efficiency of memory. Persistent memory is a disruptive technology that drastically reduces memory cost and static power but introduces the problems of slow writes and limited write endurance. An effective solution is caching. However, existing caches have been designed for fast reads: they do not minimize the number of write backs from cache to memory. The intellectual merits of this project are sufficient theories and efficient techniques for optimal write caching in future systems that use persistent memory. Caching, in both software and hardware, is increasingly prevalent. Hardware caches serve most of the memory accesses on commodity machines as well as super computers. Software caches, e.g. Memcached used by Wikipedia and Facebook, serve most of the data requests from online users. The project's broader significance and importance are the broad benefits of cache optimization in scientific and consumer computing.This project develops the formal concept of write locality and the theory and techniques to characterize locality and optimize the cache. It has three parts. The first is write locality modeling, which includes a set of metrics that can derive the amount of modified data in cache and the write backs from cache to memory in cache of all sizes and degrees of sharing. The second is deductive optimization, which mathematically derives the cache performance of different task and data organizations and selects the best solution among a combinatorial number of choices. The last is system modeling and application optimization, which use write locality to model software and hardware cache and to optimize a program.

极限计算的限制是容量，速度，耐用性，最重要的是记忆的能源效率。持续记忆是一项破坏性的技术，可大大降低记忆成本和静态力量，但引入了缓慢写入和有限的写入耐力的问题。有效的解决方案是缓存。但是，现有的缓存是为快速读取而设计的：它们不会最大程度地减少从缓存到内存的书写次数的数量。 该项目的智力优点是足够的理论和有效的技术，可以在使用持续记忆的未来系统中最佳写作缓存。 在软件和硬件中，缓存越来越普遍。硬件缓存服务在商品机和超级计算机上的大多数内存访问。软件缓存，例如Wikipedia和Facebook使用的MEMCACHED为在线用户提供了大多数数据请求。 该项目在科学和消费者计算中的较广泛的意义和重要性是缓存优化的广泛好处。该项目开发了写作局部性的正式概念以及表征局部性并优化缓存的理论和技术。 它有三个部分。 首先是写入locality建模，其中包括一组指标，这些指标可以在缓存中得出修改后的数据数量以及从缓存到内存的书面返回，以各种尺寸和共享程度的缓存。 第二个是演绎优化，它从数学上得出了不同任务和数据组织的缓存性能，并在组合数量的选择中选择了最佳解决方案。最后一个是系统建模和应用程序优化，它们使用写入局部性来建模软件和硬件缓存并优化程序。

项目成果

Timescale functions for parallel memory allocation

• DOI: 10.1145/3315573.3329987
• 发表时间: 2019-06
• 期刊: Proceedings of the 2019 ACM SIGPLAN International Symposium on Memory Management
• 作者: Pengcheng Li;Hao Luo;C. Ding

Prediction and bounds on shared cache demand from memory access interleaving

• DOI: 10.1145/3210563.3210565
• 发表时间: 2018-06
• 期刊: Proceedings of the 2018 ACM SIGPLAN International Symposium on Memory Management
• 作者: Jacob Brock;C. Ding;Rahman Lavaee;Fangzhou Liu;Liang Yuan

Locality analysis through static parallel sampling

• DOI: 10.1145/3192366.3192402
• 发表时间: 2018-06
• 期刊: Proceedings of the 39th ACM SIGPLAN Conference on Programming Language Design and Implementation
• 作者: Dong Chen;Fangzhou Liu;C. Ding;Sreepathi Pai

LD: Low-Overhead GPU Race Detection Without Access Monitoring

LD：无需访问监控的低开销 GPU 竞争检测

• DOI: 10.1145/3046678
• 发表时间: 2017
• 期刊: ACM Transactions on Architecture and Code Optimization
• 影响因子: 1.6
• 作者: Li, Pengcheng;Hu, Xiaoyu;Chen, Dong;Brock, Jacob;Luo, Hao;Zhang, Eddy Z.;Ding, Chen
• 通讯作者: Ding, Chen

Memory equalizer for lateral management of heterogeneous memory

• DOI: 10.1145/3132402.3132437
• 发表时间: 2017-10
• 期刊: Proceedings of the International Symposium on Memory Systems
• 作者: Chencheng Ye;Chen Ding;Hai Jin