Fluid Petri Nets for the Performance Evaluation of MapReduce and Spark Applications

Eugenio Gianniti; Alessandro Maria Rizzi; Enrico Barbierato; Marco Gribaudo; Danilo Ardagna

doi:10.1145/3092819.3092824

Fluid Petri Nets for the Performance Evaluation of MapReduce and Spark Applications

Eugenio Gianniti, Alessandro Maria Rizzi, Enrico Barbierato, Marco Gribaudo, Danilo Ardagna^*

^*Autore corrispondente per questo lavoro

Polytechnic University of Milan

Risultato della ricerca: Contributo in libro › Contributo a convegno

Abstract

Big Data applications allow to successfully analyze large amounts of data not necessarily structured, though at the same time they present new challenges. For example, predicting the performance of frameworks such as Hadoop can be a costly task, hence the necessity to provide models that can be a valuable support for designers and developers. This paper provides a new contribution in studying a novel modeling approach based on fluid Petri nets to predict MapReduce jobs execution time. The experiments we performed at CINECA, the Italian supercomputing center, have shown that the achieved accuracy is within 16% of the actual measurements on average.

Lingua originale	English
Titolo della pubblicazione ospite	Performance Evaluation Review
Pagine	23-36
Numero di pagine	14
Volume	44
DOI	https://doi.org/10.1145/3092819.3092824
Stato di pubblicazione	Pubblicato - 2017
Evento	10th EAI International Conference on Performance Evaluation Methodologies and Tools - Taormina Durata: 25 ott 2016 → 28 ott 2016

Convegno

Convegno	10th EAI International Conference on Performance Evaluation Methodologies and Tools
Città	Taormina
Periodo	25/10/16 → 28/10/16

Keywords

Fluid models, performance

Accesso al documento

10.1145/3092819.3092824

Altri file e collegamenti

Link a IRIS PubliCatt

Cita questo

@inproceedings{955f83e4898746f5ac8a1596d311e225,

title = "Fluid Petri Nets for the Performance Evaluation of MapReduce and Spark Applications",

abstract = "Big Data applications allow to successfully analyze large amounts of data not necessarily structured, though at the same time they present new challenges. For example, predicting the performance of frameworks such as Hadoop can be a costly task, hence the necessity to provide models that can be a valuable support for designers and developers. This paper provides a new contribution in studying a novel modeling approach based on fluid Petri nets to predict MapReduce jobs execution time. The experiments we performed at CINECA, the Italian supercomputing center, have shown that the achieved accuracy is within 16% of the actual measurements on average.",

keywords = "Fluid models, performance, Fluid models, performance",

author = "Eugenio Gianniti and Rizzi, {Alessandro Maria} and Enrico Barbierato and Marco Gribaudo and Danilo Ardagna",

year = "2017",

doi = "10.1145/3092819.3092824",

language = "English",

isbn = "978-1-4503-5846-0",

volume = "44",

pages = "23--36",

booktitle = "Performance Evaluation Review",

note = "10th EAI International Conference on Performance Evaluation Methodologies and Tools ; Conference date: 25-10-2016 Through 28-10-2016",

}

TY - GEN

T1 - Fluid Petri Nets for the Performance Evaluation of MapReduce and Spark Applications

AU - Gianniti, Eugenio

AU - Rizzi, Alessandro Maria

AU - Barbierato, Enrico

AU - Gribaudo, Marco

AU - Ardagna, Danilo

PY - 2017

Y1 - 2017

N2 - Big Data applications allow to successfully analyze large amounts of data not necessarily structured, though at the same time they present new challenges. For example, predicting the performance of frameworks such as Hadoop can be a costly task, hence the necessity to provide models that can be a valuable support for designers and developers. This paper provides a new contribution in studying a novel modeling approach based on fluid Petri nets to predict MapReduce jobs execution time. The experiments we performed at CINECA, the Italian supercomputing center, have shown that the achieved accuracy is within 16% of the actual measurements on average.

AB - Big Data applications allow to successfully analyze large amounts of data not necessarily structured, though at the same time they present new challenges. For example, predicting the performance of frameworks such as Hadoop can be a costly task, hence the necessity to provide models that can be a valuable support for designers and developers. This paper provides a new contribution in studying a novel modeling approach based on fluid Petri nets to predict MapReduce jobs execution time. The experiments we performed at CINECA, the Italian supercomputing center, have shown that the achieved accuracy is within 16% of the actual measurements on average.

KW - Fluid models, performance

UR - http://hdl.handle.net/10807/202860

U2 - 10.1145/3092819.3092824

DO - 10.1145/3092819.3092824

M3 - Conference contribution

SN - 978-1-4503-5846-0

VL - 44

SP - 23

EP - 36

BT - Performance Evaluation Review

T2 - 10th EAI International Conference on Performance Evaluation Methodologies and Tools

Y2 - 25 October 2016 through 28 October 2016

ER -

Fluid Petri Nets for the Performance Evaluation of MapReduce and Spark Applications

Abstract

Convegno

Keywords

Accesso al documento

Altri file e collegamenti

Fingerprint

Cita questo