Hadoop mapreduce scheduling paradigms

Johannessen, Roger; Yazidi, Anis; Feng, Boning

Johannessen, Roger; Yazidi, Anis; Feng, Boning

Chapter, Peer reviewed

Accepted version

Åpne

Hadoop_Scheduling_Survey_Roger_Anis_Boning.pdf (227.1Kb)

Permanent lenke

https://hdl.handle.net/10642/6436

Utgivelsesdato

2017

Metadata

Vis full innførsel

Samlinger

TKD - Institutt for informasjonsteknologi [960]

Originalversjon

Johannessen, Y.A. & Feng, B. (2017). Hadoop mapreduce scheduling paradigms. I J. Zhu, E.B. Lin & T.Li (Red.), 2017 IEEE 2nd International Conference on Cloud Computing and Big Data Analysis (ICCCBDA). Red Hook, NY: IEEE, s. 175-179

Sammendrag

Apache Hadoop is one of the most prominent and early technologies for handling big data. Different scheduling algorithms within the framework of Apache Hadoop were developed in the last decade. In this paper, we attempt to provide a comprehensive overview over the different paradigms for scheduling in Apache Hadoop. The surveyed approaches fall under different categories, namely, Deadline prioritization, Resource prioritization, Job size prioritization, Hybrid approaches and recent trends for improvements upon default schedulers.

Utgiver

Institute of Electrical and Electronics Engineers