位置:成果数据库 > 期刊 > 期刊详情页
Endurance Awarded Compiling Optimization
  • ISSN号:1672-5913
  • 期刊名称:《计算机教育》
  • 时间:0
  • 分类:TP391.41[自动化与计算机技术—计算机应用技术;自动化与计算机技术—计算机科学与技术]
  • 作者机构:School of Technology, Nanjing Audit University, Nanjing210029, Jiangsu, China
  • 相关基金:Supported by the Natural Science Foundation of Jiangsu Province(15KJB520019);High Level Talent Project of“Six Talents Summit”in Jiangsu Province;the Priority Academic Program Development Project of Jiangsu Higher Education Institution
中文摘要:

The unbanlanced updating frequency of data reduces the endurance of the whole non-volatile main memory that affects the system’s stability significantly. Wear-leveling techniques are effective methods to increase the endurance of non-volatile memory. In this paper, we propose a pure software based low cost wear-leveling method, called writing balance aware mandatory rewriting insertion optimization(WBAMRI), to improve the dynamic wear-leveling techniques.This method uses the mandatory rewriting operation combined with simple dynamic wear-leveling techniques to switch the code area to hot area dynamically. From the result of simulation, we can see that: 1) Compared with dynamic wear-leveling techniques, our ware-leveling method makes the endurance of non-volatile memory 6 times longer; 2) Compared with static wear-leveling techniques, our method nearly gets the same endurance but without the complex hardware design and the lose of memory access performance.

英文摘要:

The unbanlanced updating frequency of data reduces the endurance of the whole non-volatile main memory that affects the system’s stability significantly. Wear-leveling techniques are effective methods to increase the endurance of non-volatile memory. In this paper, we propose a pure software based low cost wear-leveling method, called writing balance aware mandatory rewriting insertion optimization(WBAMRI), to improve the dynamic wear-leveling techniques.This method uses the mandatory rewriting operation combined with simple dynamic wear-leveling techniques to switch the code area to hot area dynamically. From the result of simulation, we can see that: 1) Compared with dynamic wear-leveling techniques, our ware-leveling method makes the endurance of non-volatile memory 6 times longer; 2) Compared with static wear-leveling techniques, our method nearly gets the same endurance but without the complex hardware design and the lose of memory access performance.

同期刊论文项目
同项目期刊论文
期刊信息
  • 《计算机教育》
  • 主管单位:教育部
  • 主办单位:清华大学
  • 主编:奚春燕
  • 地址:北京市海淀区双清路清华大学学研大厦B座606室
  • 邮编:100084
  • 邮箱:jsjjy@vip.163.com
  • 电话:010-62770175-3402-3406
  • 国际标准刊号:ISSN:1672-5913
  • 国内统一刊号:ISSN:11-5006/TP
  • 邮发代号:80-171
  • 获奖情况:
  • 国内外数据库收录:
  • 被引量:26095