Request-aware cooperative I/O scheduling for scale-out database applications

Hyungil Jo, Sung hun Kim, Sangwook Kim, Jinkyu Jeong, Joonwon Lee

Research output: Contribution to conferencePaperpeer-review

2 Citations (Scopus)

Abstract

Interactive data center applications suffer from the tail latency problem. Since most modern data center applications take the sharded architecture to serve scale-out services, a request comprises multiple sub-requests handled in individual back-end nodes. Depending on the state of each back-end node, a node may issue multiple I/Os for a single sub-request. Since traditional I/O scheduling operates in an application-agnostic manner, it sometimes causes a long latency gap between the responses of sub-requests, thereby delaying the response to end-users. In this paper, we propose a request-aware cooperative I/O scheduling scheme to reduce the tail latency of a database application. Our proposed scheme captures request arrival order at the front-end of an application and exploits it to make a decision for I/O scheduling in individual back-end nodes. We implemented a prototype based on MongoDB and the Linux kernel and evaluated it with a read-intensive scan workload. Experimental results show that our proposed scheme effectively reduces the latency gap between sub-requests, thereby reducing the tail latency.

Original languageEnglish
Publication statusPublished - 2017
Event9th USENIX Workshop on Hot Topics in Storage and File Systems, HotStorage 2017, co-located with USENIX ATC 2017 - Santa Clara, United States
Duration: 2017 Jul 102017 Jul 11

Conference

Conference9th USENIX Workshop on Hot Topics in Storage and File Systems, HotStorage 2017, co-located with USENIX ATC 2017
Country/TerritoryUnited States
CitySanta Clara
Period17/7/1017/7/11

Bibliographical note

Publisher Copyright:
© 2017 USENIX Association. All rights reserved.

All Science Journal Classification (ASJC) codes

  • Hardware and Architecture
  • Information Systems
  • Software
  • Computer Networks and Communications

Fingerprint

Dive into the research topics of 'Request-aware cooperative I/O scheduling for scale-out database applications'. Together they form a unique fingerprint.

Cite this