Self-learnable cluster-based prefetching method for DRAM-flash hybrid main memory architecture

Su Kyung Yoon, Young Sun Youn, bernd Burgstaller, Shin Dug Kim

Research output: Contribution to journalArticle

Abstract

This article presents a novel prefetching mechanism for memory-intensive workloads used in large-scale data centers. We design a negative-AND-flash/dynamic random-access memory (DRAM) hybrid memory architecture as a cost-effective memory architecture to resolve the scalability and power consumption problems of a DRAM-based model. A smart prefetching mechanism based on a cluster-management scheme to cope with dynamically varying and complex access patterns of any given application is designed for maximizing the performance of the DRAM. In this article, we propose a new concept for page management, called a cluster, which prefetches data in our hybrid memory architecture. The cluster management is based on a self-learning scheme on dynamically changeable access patterns by considering any correlation between missed pages. Experimental results show that the overall performance is significantly improved in relation to hit rate, execution time, and energy consumption. Namely, our proposed model can enhance the hit rate by 15% and reduce the execution time by 1.75 times. In addition, we can save energy consumption by around 48% by cutting the number of flushed pages to about an eighth of that in a conventional system.

Original languageEnglish
Article number10
JournalACM Journal on Emerging Technologies in Computing Systems
Volume15
Issue number1
DOIs
Publication statusPublished - 2019 Jan 1

Fingerprint

Memory architecture
Flash memory
Data storage equipment
Energy utilization
Scalability
Electric power utilization
Costs

All Science Journal Classification (ASJC) codes

  • Software
  • Hardware and Architecture
  • Electrical and Electronic Engineering

Cite this

@article{609019ec4b2749c0ba25cab0c31129cb,
title = "Self-learnable cluster-based prefetching method for DRAM-flash hybrid main memory architecture",
abstract = "This article presents a novel prefetching mechanism for memory-intensive workloads used in large-scale data centers. We design a negative-AND-flash/dynamic random-access memory (DRAM) hybrid memory architecture as a cost-effective memory architecture to resolve the scalability and power consumption problems of a DRAM-based model. A smart prefetching mechanism based on a cluster-management scheme to cope with dynamically varying and complex access patterns of any given application is designed for maximizing the performance of the DRAM. In this article, we propose a new concept for page management, called a cluster, which prefetches data in our hybrid memory architecture. The cluster management is based on a self-learning scheme on dynamically changeable access patterns by considering any correlation between missed pages. Experimental results show that the overall performance is significantly improved in relation to hit rate, execution time, and energy consumption. Namely, our proposed model can enhance the hit rate by 15{\%} and reduce the execution time by 1.75 times. In addition, we can save energy consumption by around 48{\%} by cutting the number of flushed pages to about an eighth of that in a conventional system.",
author = "Yoon, {Su Kyung} and Youn, {Young Sun} and bernd Burgstaller and Kim, {Shin Dug}",
year = "2019",
month = "1",
day = "1",
doi = "10.1145/3284932",
language = "English",
volume = "15",
journal = "ACM Journal on Emerging Technologies in Computing Systems",
issn = "1550-4832",
publisher = "Association for Computing Machinery (ACM)",
number = "1",

}

Self-learnable cluster-based prefetching method for DRAM-flash hybrid main memory architecture. / Yoon, Su Kyung; Youn, Young Sun; Burgstaller, bernd; Kim, Shin Dug.

In: ACM Journal on Emerging Technologies in Computing Systems, Vol. 15, No. 1, 10, 01.01.2019.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Self-learnable cluster-based prefetching method for DRAM-flash hybrid main memory architecture

AU - Yoon, Su Kyung

AU - Youn, Young Sun

AU - Burgstaller, bernd

AU - Kim, Shin Dug

PY - 2019/1/1

Y1 - 2019/1/1

N2 - This article presents a novel prefetching mechanism for memory-intensive workloads used in large-scale data centers. We design a negative-AND-flash/dynamic random-access memory (DRAM) hybrid memory architecture as a cost-effective memory architecture to resolve the scalability and power consumption problems of a DRAM-based model. A smart prefetching mechanism based on a cluster-management scheme to cope with dynamically varying and complex access patterns of any given application is designed for maximizing the performance of the DRAM. In this article, we propose a new concept for page management, called a cluster, which prefetches data in our hybrid memory architecture. The cluster management is based on a self-learning scheme on dynamically changeable access patterns by considering any correlation between missed pages. Experimental results show that the overall performance is significantly improved in relation to hit rate, execution time, and energy consumption. Namely, our proposed model can enhance the hit rate by 15% and reduce the execution time by 1.75 times. In addition, we can save energy consumption by around 48% by cutting the number of flushed pages to about an eighth of that in a conventional system.

AB - This article presents a novel prefetching mechanism for memory-intensive workloads used in large-scale data centers. We design a negative-AND-flash/dynamic random-access memory (DRAM) hybrid memory architecture as a cost-effective memory architecture to resolve the scalability and power consumption problems of a DRAM-based model. A smart prefetching mechanism based on a cluster-management scheme to cope with dynamically varying and complex access patterns of any given application is designed for maximizing the performance of the DRAM. In this article, we propose a new concept for page management, called a cluster, which prefetches data in our hybrid memory architecture. The cluster management is based on a self-learning scheme on dynamically changeable access patterns by considering any correlation between missed pages. Experimental results show that the overall performance is significantly improved in relation to hit rate, execution time, and energy consumption. Namely, our proposed model can enhance the hit rate by 15% and reduce the execution time by 1.75 times. In addition, we can save energy consumption by around 48% by cutting the number of flushed pages to about an eighth of that in a conventional system.

UR - http://www.scopus.com/inward/record.url?scp=85061076396&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85061076396&partnerID=8YFLogxK

U2 - 10.1145/3284932

DO - 10.1145/3284932

M3 - Article

VL - 15

JO - ACM Journal on Emerging Technologies in Computing Systems

JF - ACM Journal on Emerging Technologies in Computing Systems

SN - 1550-4832

IS - 1

M1 - 10

ER -