SHF: SMALL: Collaborative Research: Improving Reliability of In-Memory Storage

SHF：SMALL：协作研究：提高内存存储的可靠性

• 批准号: 1745748
• 负责人: Jianhui Yue
• 金额: $ 17.69万
• 依托单位: Michigan Technological University
• 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: 2017-06-26 至 2022-07-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1745748&HistoricalAwards=false
• 关键词: SHF; SMALL; Collaborative; Research; Improving

Emerging nonvolatile memory (NVM) technologies, such as PCM, STT-RAM, and memristors, provide not only byte-addressability, low-latency reads and writes comparable to DRAM, but also persistent writes and potentially large storage capacity like an SSD. These advantages make NVM likely to be next-generation fast persistent storage for massive data, referred to as in-memory storage. Yet, NVM-based storage has two challenges: (1) Memory cells have limited write endurance (i.e., the total number of program/erase cycles per cell); (2) NVM has to remain in a consistent state in the event of a system crash or power loss. The goal of this project is to develop an efficient in-memory storage framework that addresses these two challenges. This project involves undergraduate and graduate students. All software artifacts and tools will be made available to the wider research community. The work has broader industrial and economic impact since it will help improve the reliability of data storage systems for data centers and HPC applications. This project will take a holistic approach, spanning from low-level architecture design to high-level OS management, to optimize the reliability, performance, and manageability of in-memory storage. The technical approach will involve understanding the implication and impact of the write endurance issue when cutting-edge NVM is adopted into storage systems. The improved understanding will motivate and aid the design of cost-effective methods to improve the life-time of in-memory storage and to achieve efficient and reliable consistence maintenance.

新兴的非挥发记忆（NVM）技术（例如PCM，STT-RAM和MEMRISTORS）不仅提供字节 - 地址性能，低延迟读取和写入与DRAM相当，而且还提供了持久的写作，而且具有SSD的潜在大型存储能力。这些优势使NVM可能是大量数据的下一代快速持续存储，称为内存存储。然而，基于NVM的存储有两个挑战：（1）记忆细胞的写入耐力有限（即，每个单元的程序/擦除周期的总数）； （2）如果系统崩溃或电力损耗，NVM必须保持一致的状态。该项目的目的是开发一个有效的内存存储框架，以解决这两个挑战。该项目涉及本科生和研究生。所有软件工件和工具都将提供给更广泛的研究社区。这项工作具有更广泛的工业和经济影响，因为它将有助于提高数据中心和HPC应用程序数据存储系统的可靠性。该项目将采用一种整体方法，从低级体系结构设计到高级操作系统管理，以优化内存存储的可靠性，性能和可管理性。技术方法将涉及理解尖端NVM被用于存储系统时的写作耐力问题的含义和影响。改善的理解将激发和帮助设计具有成本效益的方法，以改善内存存储的寿命并实现有效和可靠的一致性维护。

项目成果

Accelerate Hardware Logging for Efficient Crash Consistency in Persistent Memory

加速硬件日志记录以实现持久内存中的高效崩溃一致性

• 发表时间: 2022
• 期刊: Automation and Test in Europe Conference(DATE
• 作者: Zhiyuan Lu, Jianhui Yue

Efficient Hardware-assisted Out-place Update for Persistent Memory

高效的硬件辅助持久内存异地更新

• 发表时间: 2021
• 期刊: Automation and Test in Europe Conference(DATE
• 作者: Yifu Deng, Jianhui Yue

Efficient Hardware Redo Log for Secure Persistent Memory

用于安全持久内存的高效硬件重做日志

• 发表时间: 2021
• 期刊: 23rd International Conferences on High Performance Computing and Communications (HPCC 2021
• 作者: Zhan Zhang, Jianhui Yue

FlashWalker: An In-Storage Accelerator for Graph Random Walks

FlashWalker：用于图随机游走的存储内加速器

• 发表时间: 2022
• 期刊: 36th IEEE International Parallel & Distributed Processing Symposium(IPDPS
• 作者: Fuping Niu, Jianhui Yue

Efficient NVM Crash Consistency by Mitigating Resource Contention

通过减少资源争用实现高效的 NVM 崩溃一致性

• 发表时间: 2021
• 期刊: and Storage (NAS 2021
• 作者: Zhiyuan Lu, Jianhui Yue