采用流水化伪随机编码算法的相变存储器寿命延长方法
发布时间:2018-01-07 08:02
本文关键词:采用流水化伪随机编码算法的相变存储器寿命延长方法 出处:《计算机研究与发展》2017年06期 论文类型:期刊论文
更多相关文章: 相变存储器件 差分写入 伪随机编码算法 编码方向模式 星型生成规则
【摘要】:相变存储器(phase change memory,PCM)是一种颇具前景的新型存储器件,具有非易失性、静态功耗低和存储密度高的优点.然而,该类器件的低写入寿命是其在实用化中亟待克服的关键问题之一.一般来说,通过每次写入时仅写入相异位的策略,可以减少产生的平均写入量,从而延长PCM的写入寿命.然而,应用这一差异式的写入策略通常又会以降低读写速度为代价.为此,提出了一种兼具高效和快速特点的写入量减少方法 FEBRE(a fast and efficient bit-flipping reduction technique to extend PCM lifetime).该方法在差分写入阶段前,设计并使用了一种快速的一对多映射,将待写入的数据并行映射为多个编码向量,从而增加了从其中找到一个与已有数据最近的向量的可能性.此外,还提出了一种流水化的伪随机编码算法,用以加速一对多映射中的编码过程,从而降低写入开销.实验表明,与目前领先的PRES(pseudo-random encoding scheme)方法相比,FEBRE方法在写入操作中,平均减少了5%以上的写入量,提升了2倍以上的编码速度;在读取操作中,减少了45%以上的解码操作次数.
[Abstract]:Phase change memory (change) is a promising new type of memory device with the advantages of non-volatile, low static power consumption and high storage density. The low writing lifetime of this kind of devices is one of the key problems that need to be overcome in practical application. Generally speaking, the average amount of writing can be reduced by the strategy of only writing ectopic in each write. This extends the write life of PCM. However, the application of this differential write strategy usually comes at the expense of reading and writing speed. In this paper, a fast and efficient write reduction method, FEBREE, is proposed. A fast and efficient bit-flipping reduction technique to extend PCM. Lifetime). The method is prior to the differential write phase. A fast one-to-many mapping is designed and used to map the data to multiple encoding vectors in parallel, which increases the possibility of finding a vector closest to the existing data. A income pseudorandom coding algorithm is also proposed to speed up the coding process in one-to-many mapping, thus reducing the write overhead. Compared with the current leading PRES(pseudo-random encoding schememethod, the FEBRE method is in the write operation. The average amount of writing is reduced by more than 5%, and the coding speed is increased by more than 2 times. In the read operation, the number of decoding operations is reduced by more than 45%.
【作者单位】: 清华大学计算机科学与技术系;清华大学信息技术研究院;
【基金】:国家重点研发计划(2016YFB1000303) 国家自然科学基金项目(61373025)~~
【分类号】:TP333
【正文快照】: This work was supported by the National Key Research and Development Plan of China(2016YFB1000303)and the NationalNatural Science Foundation of China(61373025).(gaopeng1982@gmail.com)随着软硬件技术的发展,现代计算机系统对高速、大容量、低功耗的内存系统产生
【相似文献】
相关期刊论文 前3条
1 李世军;;JPEG图像压缩编码算法的DSP优化实现[J];微计算机信息;2012年09期
2 胡国兴;沈海斌;;低翻转率的SoC总线组合编码算法[J];电子器件;2006年04期
3 ;[J];;年期
相关硕士学位论文 前2条
1 王全明;H.264编码算法在双核DSP上的实现方法研究[D];华中科技大学;2007年
2 葛力雅;AC-3编码算法研究与DSP上的实现[D];天津大学;2007年
,本文编号:1391665
本文链接:https://www.wllwen.com/kejilunwen/jisuanjikexuelunwen/1391665.html