Evaluating Lifetime of Server Data based on Trace Analysis

Minhyun Park, Kyungah Lee, Ilhoon Shin

Research output: Contribution to journalArticlepeer-review

1 Scopus citations

Abstract

SSD (Solid state drives) performance can be improved by applying a shallow write to short-lived data. For this, the lifetime of the data should be accurately predicted, and different NAND write mode should be applied according to the data lifetime. A wrong prediction can result in degrading the performance and the lifetime of SSDs. This study analyses the factors that affect the lifetime of the data and presents policies that predict the lifetime of the data more accurately. The presented policies improves the accuracy by 2.2-89.1% compared to the existing policies by considering both write request size and previous lifetime of the data.

Original languageEnglish
Pages (from-to)1441-1444
Number of pages4
JournalInternational Journal of Engineering Research and Technology
Volume12
Issue number9
StatePublished - 2019

Keywords

  • Lifetime of data
  • Server workload
  • SSD

Fingerprint

Dive into the research topics of 'Evaluating Lifetime of Server Data based on Trace Analysis'. Together they form a unique fingerprint.

Cite this