Wear-leveling method for PCM based on random mapping

Wei Liu, Yuping Wang*

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

Abstract

Phase change memory (PCM) uses a phase change material that is non-volatile, higher density, lower energy, and has better scalability than other methods. The main problem with PCM materials is the cell write limitation. Recent studies have focused on optimizing the write operations and wear-leveling, but cannot prevent malicious attacks to rapidly wear the PCM cells. This paper presents a PCM wear-leveling method which divides the whole PCM memory into two levels with separate random mapping tables to convert the logical cell addresses into physical addresses. The random mapping tables are updated dynamically according to the write count thresholds. This method not only implements PCM wear-leveling, but also resists malicious wearing-out attacks. Tests show that This method improves the wear-leveling by up to 87.5% over three existing wear-leveling methods with a system performance loss of less than 6% and additional storage overhead of less than 1‰ of the whole memory size.

Original languageEnglish
Pages (from-to)1208-1215
Number of pages8
JournalQinghua Daxue Xuebao/Journal of Tsinghua University
Volume55
Issue number11
DOIs
Publication statusPublished - 1 Nov 2015
Externally publishedYes

Keywords

  • Operating system
  • Phase change memory
  • Random mapping
  • Wear-leveling

Fingerprint

Dive into the research topics of 'Wear-leveling method for PCM based on random mapping'. Together they form a unique fingerprint.

Cite this