首页> 外文会议>Advanced parallel processing technologies >A Read-Write Aware Replacement Policy for Phase Change Memory
【24h】

A Read-Write Aware Replacement Policy for Phase Change Memory

机译:相变存储器的读写感知替换策略

获取原文
获取原文并翻译 | 示例

摘要

Scaling DRAM will be increasingly di.cult due to power and cost constraint. Phase Change Memory (PCM) is an emerging memory technology that can increase main memory capacity in a cost-e.ective and power-e.cient manner. However, PCM incurs relatively long latency , high write energy, and .nite endurance. To make PCM an alternative for scalable main memory, write tra.c to PCM should be reduced, where memory replacement policy could play a vital role. In this paper, we propose a Read-Write Aware policy (RWA) to reduce write tra.c without performance degradation. RWA explores the asymmetry of read and write costs of PCM, and prevents dirty data lines from frequent evictions. Simulations results on an 8-core CMP show that for memory organization with and without DRAM bu.er, RWA can achieve 33.1% and 14.2% reduction in write tra.c to PCM respectively. In addition, an Improved RWA (I-RWA) is proposed that takes into consideration the write access pattern and can further improve memory e.ciency. For organization with DRAM bu.er, I-RWA provides a signi.cant 42.8% reduction in write tra.c. Furthermore, both RWA and I-RWA incurs no hardware overhead and can be easily integrated into existing hardware.
机译:由于功率和成本的限制,扩展DRAM将越来越困难。相变存储器(PCM)是一种新兴的存储器技术,可以以节省成本和节省功率的方式增加主存储器的容量。但是,PCM会导致较长的等待时间,较高的写入能量和耐久力。为了使PCM成为可扩展主存储器的替代方案,应减少对PCM的tra.c写入,在这种情况下,存储器替换策略可能起着至关重要的作用。在本文中,我们提出了一种读写感知策略(RWA)以减少写tra.c而不会降低性能。 RWA探索PCM读写成本的不对称性,并防止脏数据线频繁迁离。在8核CMP上进行的仿真结果表明,对于有和没有DRAM缓冲区的内存组织,RWA可以分别将PCM的写入流量减少33.1%和14.2%。此外,提出了一种改进的RWA(I-RWA),它考虑了写访问模式并可以进一步提高存储效率。对于具有DRAM缓冲区的组织,I-RWA可将写入流量显着降低42.8%。此外,RWA和I-RWA都没有硬件开销,并且可以轻松集成到现有硬件中。

著录项

  • 来源
  • 会议地点 Shanghai(CN)
  • 作者单位

    Tsinghua National Laboratory for Information Science and Technology Department of Computer Science Technology, Tsinghua University, Beijing 100084, China;

    Tsinghua National Laboratory for Information Science and Technology Department of Computer Science Technology, Tsinghua University, Beijing 100084, China;

    Tsinghua National Laboratory for Information Science and Technology Department of Computer Science Technology, Tsinghua University, Beijing 100084, China;

    Tsinghua National Laboratory for Information Science and Technology Department of Computer Science Technology, Tsinghua University, Beijing 100084, China;

    Tsinghua National Laboratory for Information Science and Technology Department of Computer Science Technology, Tsinghua University, Beijing 100084, China;

  • 会议组织
  • 原文格式 PDF
  • 正文语种 eng
  • 中图分类 理论、方法;
  • 关键词

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号