计算机科学 ›› 2018, Vol. 45 ›› Issue (11A): 505-510.

• 软件工程与数据库技术 • 上一篇    下一篇

带磨损均衡的小粒度非易失性内存管理机制

孙强1, 诸葛晴凤2, 陈咸彰1,3, 沙行勉1,2, 吴林1   

  1. 重庆大学计算机学院 重庆4000441
    华东师范大学计算机科学与软件工程学院 上海2000622
    重庆大学通信工程学院 重庆4000443
  • 出版日期:2019-02-26 发布日期:2019-02-26
  • 通讯作者: 陈咸彰(1989-),男,博士,讲师,CCF会员,主要研究领域为新型内存系统、文件系统、嵌入式系统和软件、云计算,E-mail:xzchen@cqu.edu.cn
  • 作者简介:孙 强(1992-),男,硕士生,主要研究领域为新型内存系统、内存数据库;诸葛晴凤(1970-),女,博士,教授,博士生导师,CCF会员,主要研究领域为操作系统、嵌入式系统和软件、优化算法;沙行勉(1964-),男,博士,教授,博士生导师,CCF高级会员,主要研究领域为新型内存系统、操作系统、云计算和嵌入式系统和软件;吴 林(1991-),男,博士生,主要研究领域为新型内存系统、文件系统。
  • 基金资助:
    本文受国家自然科学基金(61472052),中国博士后科学基金(2017M620412)资助。

In-page Wear-leveling Memory Management Based on Non-volatile Memory

SUN Qiang1, ZHUGE Qing-feng2, CHEN Xian-zhang1,3, Edwin H.-M.SHA1,2, WU Lin1   

  1. College of Computer Science,Chongqing University,Chongqing 400044,China1
    School of Computer Science and Software Engineering,East China Normal University,Shanghai 200062,China2
    College of Communication Engineering,Chongqing University,Chongqing 400044,China3
  • Online:2019-02-26 Published:2019-02-26

摘要: 非易失性内存以其卓越的特性被视作具有巨大潜力的下一代存储设备。然而,非易失性存储单元存在写耐受度低的缺点,使其难以承受频繁的小粒度数据更新操作。文中针对非易失性存储器,提出带磨损均衡的小粒度内存分配管理系统(IWMM)。IWMM将单个内存页分割为多个基本存储单元以适应小粒度的内存分配和数据更新操作。IWMM采用定向顺序分配算法轮流地使用单个内存页中的基本存储单元,从而将小粒度写操作均衡地分布到内存页内的各个存储单元中。实验表明,对比同样致力于磨损均衡的小粒度内存管理系统NVMalloc,IWMM能将内存页的写次数降低52.6%;同时,在内存回收率高于50%的应用场景中,性能比glibc malloc高27.6%。

关键词: 非易失性内存, 磨损均衡, 内存管理, 小粒度

Abstract: Emerging non-volatile memory(NVM) is the promising next-generation storage for its advanced characteristics.However,the low endurance of NVM cells makes them vulnerable to frequent fine-grained data updates.This paper proposed a novel in-page wear-leveling memory management (IWMM) for NVM-based storage.IWMM divides pages into basic memory units to support fine-grained updates.IWMM alternatively allocates the memory units of a page with directional order allocation algorithm to distribute fine-grained updates evenly on memory cells.Experimental results show that the wear counts of IWMM can achieve 52.6% reduction over that of NVMalloc,a wear-conscious allocator.Meanwhile,the preformance of IWMM is 27.6% betterthan glibc malloc when the ratio of memory deallocation is less than 50%.

Key words: Fine-grained, Memory management, Non-volatile memory, Wear leveling

中图分类号: 

  • TP316.89
[1]BEZ R,PIROVANO A.Non-volatile memory technologies:emerging concepts and new materials[J].Materials Science in Semiconductor Processing,2004,7(4-6):349-355.
[2]HUANG F,FENG D,HUA Y,et al.A wear-leveling-aware counter mode for data encryption in non-volatile memories[C]∥Proceedings of the Conference on Design,Automation & Test in Europe. European Design and Automation Association,2017:910-913.
[3]CHEN T Y,CHANG Y H,CHEN S H,et al.Enabling write-reduction strategy for journaling file systems over byte-addressable NVRAM[C]∥2017 54th ACM/EDAC/IEEE Design Automation Conference (DAC).IEEE,2017:1-6.
[4]SHA E H M,CHEN X,ZHUGE Q,et al.A new design of in-memory file system based on file virtual address framework[J].IEEE Transactions on Computers,2016,65(10):2959-2972.
[5]张鸿斌,范捷,舒继武,等.基于相变存储器的存储系统与技术综述[J].计算机研究与发展,2014,51(8):1647-1662.
[6]https://newsroom.intel.com/news-releases/intel-and-micron-pro-duce-breakthrough-memory-technology.
[7]EILERT S,LEINWANDER M, MCRISENZA G.Phase Change Memory (PCM):A New Memory Technology to Enable New Memory Usage Models[C]∥IEEE International Memory Workshop.2009:1-2.
[8]CONDIT J,NIGHTINGALE E B,FROST C,et al.Better i/o through byte-addressable,persistent memory[C]∥ACM SOSP.2009:133-146.
[9]DULLOOR S R,KUMAR S,KESHAVAMURTHY A,et al. System software for persistent memory[C]∥EuroSys.2014:1-15.
[10]ARULRAJ J,PAVLO A,DULLOOR S R.Let’s talk about storage & recovery methods for non-volatile memory database systems[C]∥Proceedings of the 2015 ACM SIGMOD International Conference on Management of Data.ACM,2015:707-722.
[11]RODEH O.B-trees,shadowing,and clones[J].ACM Transac-tions on Storage,2008,3(4):1-27.
[12]MOHAN C.Repeating history beyond aries[C]∥VLDB.1999:1-17.
[13]http://redis.io.
[14]SCHWALB D,BERNING T,FAUST M,et al.Nvmalloc:Me-mory allocation for nvram[C]∥AWMDS@VLDB.2015:61-72.
[15]YU S,XIAO N,DENG M,et al.Redesign the memory allocator for non-volatile main memory[J].ACM Journal on Emerging Technologies in Computing Systems(JETC),2017,13(3):49.
[16]BARKER S,MISHRA A,IRWIN D,et al.Smart*:An open data set and tools for enabling research in sustainable homes[C]∥SustKDD.2012.
[17]Umass trace repository.smart* data set for sustainability [OL].http://traces.cs.umass.edu/index.php/Smart/Smart.
[18]CHEN T Y,CHANG Y H,CHEN S H,et al.Enabling write-reduction strategy for journaling file systems over byte-addressable nvram[C]∥DAC.2017:1-6.
[19]DULLOOR S R,KUMAR S,KESHAVAMURTHY A,et al. System software for persistent memory[C]∥EuroSys.2014:1-15.
[20]SHA E H M,CHEN X,ZHUGE Q,et al.A new design of in-memory file system based on file virtual address framework[J].IEEE Transactions on Computers,2016,65(10):2959-2972.
[21]DULLOOR S R,KUMAR S,KESHAVAMURTHY A,et al. System software for persistent memory[C]∥Proceedings of the Ninth European Conference on Computer Systems.ACM,2014:15.
[22]EVANS J.A scalable concurrent glibcmalloc(3) implementation for freebsd[OL].http://www.bsdcan.org/2006/papers/jemalloc.pdf.
[23]BERGER E D,MCKINLEY K S,BLUMOFE R D,et al.Hoard:A scalable memory allocator for multithreaded applications[C]∥ASPLOS IX.2000:117-128.
[24]BHANDARI K,CHAKRABARTI D R,BOEHM H J.Makalu:Fastrecoverable allocation of non-volatile memory[C]∥ACM OOP-SLA.2016:677-694.
[1] 刘高聪, 罗永平, 金培权.
基于热点数据的持久性内存索引查询加速
Accelerating Persistent Memory-based Indices Based on Hotspot Data
计算机科学, 2022, 49(8): 26-32. https://doi.org/10.11896/jsjkx.210700176
[2] 范鹏浩, 黄国锐, 金培权.
NVRC:一种面向NVM的写限制日志方案
NVRC:Write-limited Logging for Non-volatile Memory
计算机科学, 2021, 48(3): 130-135. https://doi.org/10.11896/jsjkx.200900071
[3] 王鑫鑫, 诸葛晴凤, 吴林.
面向非易失性内存文件系统的NVM模拟与验证方法
Method for Simulating and Verifying NVM-based In-memory File Systems
计算机科学, 2020, 47(9): 74-80. https://doi.org/10.11896/jsjkx.190700037
[4] 侯泽毅, 万虎, 徐远超.
NMST:一种基于线段树的持久性内存管理优化方法
NMST:A Persistent Memory Management Optimization Approach Based on Segment Tree
计算机科学, 2018, 45(7): 78-83. https://doi.org/10.11896/j.issn.1002-137X.2018.07.012
[5] 李月,王芳.
基于NVM的存储安全综述
Survey on Storage Security of Emerging Non-volatile Memory
计算机科学, 2018, 45(7): 53-60. https://doi.org/10.11896/j.issn.1002-137X.2018.07.008
[6] 孟红涛,余松平,刘芳,肖侬.
Spark内存管理及缓存策略研究
Research on Memory Management and Cache Replacement Policies in Spark
计算机科学, 2017, 44(6): 31-35. https://doi.org/10.11896/j.issn.1002-137X.2017.06.005
[7] 余宽隆,陈渝,茅俊杰,张磊.
ARM-MuxOS:一台手机,多个世界
ARM-MuxOS:A System Architecture to Support Multiple Operating Systems on Single Mobile Device
计算机科学, 2014, 41(10): 7-11. https://doi.org/10.11896/j.issn.1002-137X.2014.10.002
[8] 吴立军,骆翔宇,陈清亮.
基于动态内存和状态管理的模型检测新方法
New Approach of Model Checking Based on Management for Dynamic Memory and State
计算机科学, 2011, 38(11): 191-195.
[9] 赵慧斌 李小群 叶以民.
Linux进程管理体系的研究与分析

计算机科学, 2003, 30(10): 84-88.
[10] 李小群 孙玉芳.
Linux内存管理机制的分析与研究

计算机科学, 2002, 29(4): 18-20.
[11] 秦莹 陈虎.
Java芯片系统存储管理的设计和实现

计算机科学, 2000, 27(3): 43-46.
[12] 陈宇 熊光泽.
基于LINUX的硬实时计算

计算机科学, 2000, 27(11): 17-19.
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
No Suggested Reading articles found!