%0 Conference Proceedings
%T FLEX: A Slot Allocation Scheduling Optimizer for MapReduce Workloads
%+ IBM Watson Research Center
%+ IBM Almaden Research Center
%A Wolf, Joel
%A Rajan, Deepak
%A Hildrum, Kirsten
%A Khandekar, Rohit
%A Kumar, Vibhore
%A Parekh, Sujay
%A Wu, Kun-Lung
%A Balmin, Andrey
%< avec comité de lecture
%( Lecture Notes in Computer Science
%B ACM/IFIP/USENIX 11th International Middleware Conference (MIDDLEWARE)
%C Bangalore, India
%Y Indranil Gupta; Cecilia Mascolo
%I Springer
%3 Middleware 2010
%V LNCS-6452
%P 1-20
%8 2010-11-29
%D 2010
%R 10.1007/978-3-642-16955-7_1
%K MapReduce
%K Scheduling
%K Allocation
%K Optimization
%Z Computer Science [cs]/Digital Libraries [cs.DL]Conference papers
%X Originally, MapReduce implementations such as Hadoop employed First In First Out (fifo) scheduling, but such simple schemes cause job starvation. The Hadoop Fair Scheduler (hfs) is a slot-based MapReduce scheme designed to ensure a degree of fairness among the jobs, by guaranteeing each job at least some minimum number of allocated slots. Our prime contribution in this paper is a different, flexible scheduling allocation scheme, known as flex. Our goal is to optimize any of a variety of standard scheduling theory metrics (response time, stretch, makespan and Service Level Agreements (slas), among others) while ensuring the same minimum job slot guarantees as in hfs, and maximum job slot guarantees as well. The flex allocation scheduler can be regarded as an add-on module that works synergistically with hfs. We describe the mathematical basis for flex, and compare it with fifo and hfs in a variety of experiments.
%G English
%2 https://inria.hal.science/hal-01055274/document
%2 https://inria.hal.science/hal-01055274/file/middleware10-cr.pdf
%L hal-01055274
%U https://inria.hal.science/hal-01055274
%~ IFIP-LNCS
%~ IFIP
%~ IFIP-LNCS-6452
%~ IFIP-MIDDLEWARE
%~ IFIP-2010