Stateful load balancing for parallel stream processing

Publikation: Bidrag til bog/antologi/rapportKonferencebidrag i proceedingsForskningfagfællebedømt

Timely processing of streams in parallel requires dynamic load balancing to diminish skewness of data. In this paper we study this problem for stateful operators with key grouping for which the process of load balancing involves a lot of state movements. Consequently, load balancing is a bi-objective optimization problem, namely Minimum-Cost-Load-Balance (MCLB). We address MCLB with two approximate algorithms by a certain relaxation of the objectives: (1) a greedy algorithm ELB performs load balancing eagerly but relaxes the objective of load imbalance to a range; and (2) a periodic algorithm CLB aims at reducing load imbalance via a greedy procedure of minimizing the covariance of substreams but ignores the objective of state movement by amortizing the overhead of it over a relative long period. We evaluate our approaches with both synthetic and real data. The results show that they can adapt effectively to load variations and improve latency efficiently comparing to the existing solutions whom ignored the overhead of state movement in stateful load balancing.
OriginalsprogEngelsk
TitelEuro-Par 2017 : Parallel Processing Workshops
RedaktørerDora B. Heras, Luc Bougé, Gabriele Mencagli, Emmanuel Jeannot, Rizos Sakellariou, Rosa M. Badia, Jorge G. Barbosa, Laura Ricci, Stephen L. Scott, Stefan Lankes, Josef Weidendorfer
Antal sider14
ForlagSpringer
Publikationsdato1 jan. 2018
Sider80-93
ISBN (Trykt)9783319751771
DOI
StatusUdgivet - 1 jan. 2018
BegivenhedInternational Workshops on Parallel Processing, Euro-Par 2017 - Santiago de Compostela, Spanien
Varighed: 28 aug. 201729 aug. 2017

Konference

KonferenceInternational Workshops on Parallel Processing, Euro-Par 2017
LandSpanien
BySantiago de Compostela
Periode28/08/201729/08/2017
NavnLecture notes in computer science
Vol/bind10659
ISSN0302-9743

ID: 194909417