A self-learning pattern adaptive prefetching method for big data applications

Xian Shu Li, Su Kyung Yoon, Jung Geun Kim, Shin-Dug Kim

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

In this study, we designed a KM-Cluster-based pattern adaptive prefetching mechanism for the last-level cache structure to support real time big data management. The goal is to predict future memory access patterns aggressively and accurately through a new self-learning prefetching engine model. The pattern adaptive last-level cache consisted of three major parts: the last-level cache, the first-level prefetching buffer (FLPB) and the second-level prefetching buffer (SLPB). The SLPB efficiently manages the history records of cache blocks evicted from the last-level cache through a self-learning mechanism. A K-means clustering algorithm is used as an SLPB prefetching scheme. Hybrid main memory is constructed using a small portion of the DRAM buffer space and primarily NAND-Flash memory space. The overall performance of our proposed model is evaluated for OpenStack Swift and in-memory database application-Redis. Experimental results show that the proposed architecture reduces the total execution time by 20.96% and power consumption by 31.9% compared to the same last-level cache size with no SLPB structure.

Original languageEnglish
Pages (from-to)66-75
Number of pages10
JournalSustainable Computing: Informatics and Systems
Volume20
DOIs
Publication statusPublished - 2018 Dec 1

Fingerprint

Data storage equipment
Flash memory
Dynamic random access storage
Clustering algorithms
Information management
Electric power utilization
Engines
Big data

All Science Journal Classification (ASJC) codes

  • Computer Science(all)
  • Electrical and Electronic Engineering

Cite this

@article{9066e60de7124a33910dff9fc9830de1,
title = "A self-learning pattern adaptive prefetching method for big data applications",
abstract = "In this study, we designed a KM-Cluster-based pattern adaptive prefetching mechanism for the last-level cache structure to support real time big data management. The goal is to predict future memory access patterns aggressively and accurately through a new self-learning prefetching engine model. The pattern adaptive last-level cache consisted of three major parts: the last-level cache, the first-level prefetching buffer (FLPB) and the second-level prefetching buffer (SLPB). The SLPB efficiently manages the history records of cache blocks evicted from the last-level cache through a self-learning mechanism. A K-means clustering algorithm is used as an SLPB prefetching scheme. Hybrid main memory is constructed using a small portion of the DRAM buffer space and primarily NAND-Flash memory space. The overall performance of our proposed model is evaluated for OpenStack Swift and in-memory database application-Redis. Experimental results show that the proposed architecture reduces the total execution time by 20.96{\%} and power consumption by 31.9{\%} compared to the same last-level cache size with no SLPB structure.",
author = "Li, {Xian Shu} and Yoon, {Su Kyung} and Kim, {Jung Geun} and Shin-Dug Kim",
year = "2018",
month = "12",
day = "1",
doi = "10.1016/j.suscom.2017.12.003",
language = "English",
volume = "20",
pages = "66--75",
journal = "Sustainable Computing: Informatics and Systems",
issn = "2210-5379",
publisher = "Elsevier USA",

}

A self-learning pattern adaptive prefetching method for big data applications. / Li, Xian Shu; Yoon, Su Kyung; Kim, Jung Geun; Kim, Shin-Dug.

In: Sustainable Computing: Informatics and Systems, Vol. 20, 01.12.2018, p. 66-75.

Research output: Contribution to journalArticle

TY - JOUR

T1 - A self-learning pattern adaptive prefetching method for big data applications

AU - Li, Xian Shu

AU - Yoon, Su Kyung

AU - Kim, Jung Geun

AU - Kim, Shin-Dug

PY - 2018/12/1

Y1 - 2018/12/1

N2 - In this study, we designed a KM-Cluster-based pattern adaptive prefetching mechanism for the last-level cache structure to support real time big data management. The goal is to predict future memory access patterns aggressively and accurately through a new self-learning prefetching engine model. The pattern adaptive last-level cache consisted of three major parts: the last-level cache, the first-level prefetching buffer (FLPB) and the second-level prefetching buffer (SLPB). The SLPB efficiently manages the history records of cache blocks evicted from the last-level cache through a self-learning mechanism. A K-means clustering algorithm is used as an SLPB prefetching scheme. Hybrid main memory is constructed using a small portion of the DRAM buffer space and primarily NAND-Flash memory space. The overall performance of our proposed model is evaluated for OpenStack Swift and in-memory database application-Redis. Experimental results show that the proposed architecture reduces the total execution time by 20.96% and power consumption by 31.9% compared to the same last-level cache size with no SLPB structure.

AB - In this study, we designed a KM-Cluster-based pattern adaptive prefetching mechanism for the last-level cache structure to support real time big data management. The goal is to predict future memory access patterns aggressively and accurately through a new self-learning prefetching engine model. The pattern adaptive last-level cache consisted of three major parts: the last-level cache, the first-level prefetching buffer (FLPB) and the second-level prefetching buffer (SLPB). The SLPB efficiently manages the history records of cache blocks evicted from the last-level cache through a self-learning mechanism. A K-means clustering algorithm is used as an SLPB prefetching scheme. Hybrid main memory is constructed using a small portion of the DRAM buffer space and primarily NAND-Flash memory space. The overall performance of our proposed model is evaluated for OpenStack Swift and in-memory database application-Redis. Experimental results show that the proposed architecture reduces the total execution time by 20.96% and power consumption by 31.9% compared to the same last-level cache size with no SLPB structure.

UR - http://www.scopus.com/inward/record.url?scp=85039694879&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85039694879&partnerID=8YFLogxK

U2 - 10.1016/j.suscom.2017.12.003

DO - 10.1016/j.suscom.2017.12.003

M3 - Article

VL - 20

SP - 66

EP - 75

JO - Sustainable Computing: Informatics and Systems

JF - Sustainable Computing: Informatics and Systems

SN - 2210-5379

ER -