.png){kind=icon}
.png){kind=icon}
Return
LPCMsim: A Lightweight Phase Change Memory Simulator
.png){kind=icon}
PRE
AI
DOI：10.1016/j.future.2019.03.028.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Phase Change Memory (PCM) is a kind of non-volatile byte-addressable memories with the potential to replace DRAM as memory, which has been widely researched. PCM has the disadvantage that too many writes can cause wear-out of PCM cells. Wear-leveling and error correction are two methods to improve the lifetime of PCM. However, performing such research with real PCM devices is time-consuming and expensive. The state-of-the-art simulators do not directly support PCM wear-leveling and error correction algorithms simulation. It is necessary to offer the community a simulator to model wear-leveling and error correction schemes for PCM. To fill the gap, this paper presents a lightweight PCM simulator named LPCMsim, which focuses on simulating wear-leveling and error correction, and does not simulate unessential functions for PCM (such as cache and network functions). To accelerate a simulating process, LPCMsim directly simulates memory accessing during simulating wear-leveling, and simulates error correction based on the normal distribution of lifetime. In order to simplify the integration of different algorithms into the simulator, LPCMsim is divided into the framework and simulated algorithms which are implemented in separated modules and are connected by an interface. LPCMsim facilitates the designers of new algorithms to validate their schemes by exploiting the exported APIs, and further, for some common algorithms, performance optimizations have already been integrated in advance. The experimental results show that LPCMsim is an efficient lightweight PCM simulator. (C) 2019 Elsevier B.V. All rights reserved.
Keywords:
Error correction
Lightweight
Phase Change Memory
Simulator
Wear-leveling
AI Summary
Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.
Journal
F
IF:
6.1
Papers:
6.8K
Citations:
2.3W