D2P: A distributed deadline propagation approach to tolerate long-tail latency in datacenters

Rui Ren, Jiuyue Ma, Xiufeng Sui, Yungang Bao

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

2 Citations (Scopus)

Abstract

We propose a Distributed Deadline Propagation (D2P) approach for datacenter applications to tolerate latency variability. The key idea of D 2P is to allow local nodes to perceive global deadline information and to propagate the information among distributed nodes. Local nodes can leverage the information to do scheduling and adjust processing speed to reduce latency variability. Preliminary experimental results show that D2P has the potential of reducing the long-tail latency in datacenters by lever-aging propagated deadline information on the local nodes.

Original languageEnglish
Title of host publicationProceedings of 5th Asia-Pacific Workshop on Systems, APSYS 2014
PublisherAssociation for Computing Machinery
ISBN (Print)9781450330244
DOIs
Publication statusPublished - 2014
Externally publishedYes
Event5th ACM Asia-Pacific Workshop on Systems, APSYS 2014 - Beijing, China
Duration: 25 Jun 201426 Jun 2014

Publication series

NameProceedings of 5th Asia-Pacific Workshop on Systems, APSYS 2014

Conference

Conference5th ACM Asia-Pacific Workshop on Systems, APSYS 2014
Country/TerritoryChina
CityBeijing
Period25/06/1426/06/14

Fingerprint

Dive into the research topics of 'D2P: A distributed deadline propagation approach to tolerate long-tail latency in datacenters'. Together they form a unique fingerprint.

Cite this