Efficient memory page management for NVDIMM-based big data processing environments

Seog Min Kwon, Hyokyung Bahn

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

1 Scopus citations

Abstract

Due to the recent computing paradigm shift from desktop to cloud environments, NVDIMM-based memory architectures for big data processing are catching considerable interest. In this paper, we present an efficient page management scheme for large-scale NVDIMM memory. The proposed architecture carefully determines the page size of a large-scale memory system such that TLB hit ratio is improved significantly without increasing the page fault ratio, thereby allowing big-data processing efficiently, which was not easy in conventional DRAM-based memory systems. The main contribution of this paper is that it proposes a new paradigm of "NVDIMM-based large memory architecture" instead of conventional "DRAM-based memory system" and presents efficient management techniques for that architecture by considering both TLB performances and page fault rates.

Original languageEnglish
Title of host publicationProceedings - 2017 4th International Conference on Information Science and Control Engineering, ICISCE 2017
EditorsYing Dai, Shaozi Li, Yun Cheng
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages283-287
Number of pages5
ISBN (Electronic)9781538630136
DOIs
StatePublished - 14 Nov 2017
Event4th International Conference on Information Science and Control Engineering, ICISCE 2017 - Changsha, Hunan, China
Duration: 21 Jul 201723 Jul 2017

Publication series

NameProceedings - 2017 4th International Conference on Information Science and Control Engineering, ICISCE 2017

Conference

Conference4th International Conference on Information Science and Control Engineering, ICISCE 2017
Country/TerritoryChina
CityChangsha, Hunan
Period21/07/1723/07/17

Bibliographical note

Publisher Copyright:
© 2017 IEEE.

Keywords

  • Big data
  • NVDIMM
  • Page fault
  • Page management
  • TLB

Fingerprint

Dive into the research topics of 'Efficient memory page management for NVDIMM-based big data processing environments'. Together they form a unique fingerprint.

Cite this