Scaling up analogy-based software effort estimation: A Comparison of multiple hadoop implementation schemes

Passakorn Phannachitta, Jacky Keung, Akito Monden, Kenichi Matsumoto

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Analogy-based estimation (ABE) is one of the most time consuming and compute intensive method in software de- velopment effort estimation. Optimizing ABE has been a dilemma because simplifying the procedure can reduce the estimation performance, while increasing the procedure com- plexity with more sophisticated theory may sacrifice an ad- vantage of the unlimited scalability for a large data input. Motivated by an emergence of cloud computing technology in software applications, in this study we present 3 different implementation schemes based on Hadoop MapReduce to optimize the ABE process across multiple computing in- stances in the cloud-computing environment. We experimentally compared the 3 MapReduce implementation schemes in contrast with our previously proposed GPGPU approach (named ABE-CUDA) over 8 high-performance Amazon EC2 instances. Results present that the Hadoop solution can pro- vide more computational resources that can extend the scalability of the ABE process. We recommend adoption of 2 different Hadoop implementations (Hadoop streaming and RHadoop) for accelerating the computation specifically for compute-intensive software engineering related tasks.

Original languageEnglish
Title of host publicationInternational Workshop on Innovative Software Development Methodologies and Practices, InnoSWDev 2014 - Proceedings
PublisherAssociation for Computing Machinery, Inc
Pages65-72
Number of pages8
ISBN (Print)9781450332262
DOIs
Publication statusPublished - Nov 16 2014
Externally publishedYes
EventInternational Workshop on Innovative Software Development Methodologies and Practices, InnoSWDev 2014 - Hong Kong, China
Duration: Nov 16 2014 → …

Other

OtherInternational Workshop on Innovative Software Development Methodologies and Practices, InnoSWDev 2014
CountryChina
CityHong Kong
Period11/16/14 → …

Fingerprint

Cloud computing
Scalability
Software engineering
Application programs

Keywords

  • Analogy-based estimation
  • Cloud computing
  • CUDA
  • Map reduce
  • Software effort estimation

ASJC Scopus subject areas

  • Software

Cite this

Phannachitta, P., Keung, J., Monden, A., & Matsumoto, K. (2014). Scaling up analogy-based software effort estimation: A Comparison of multiple hadoop implementation schemes. In International Workshop on Innovative Software Development Methodologies and Practices, InnoSWDev 2014 - Proceedings (pp. 65-72). Association for Computing Machinery, Inc. https://doi.org/10.1145/2666581.2666582

Scaling up analogy-based software effort estimation : A Comparison of multiple hadoop implementation schemes. / Phannachitta, Passakorn; Keung, Jacky; Monden, Akito; Matsumoto, Kenichi.

International Workshop on Innovative Software Development Methodologies and Practices, InnoSWDev 2014 - Proceedings. Association for Computing Machinery, Inc, 2014. p. 65-72.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Phannachitta, P, Keung, J, Monden, A & Matsumoto, K 2014, Scaling up analogy-based software effort estimation: A Comparison of multiple hadoop implementation schemes. in International Workshop on Innovative Software Development Methodologies and Practices, InnoSWDev 2014 - Proceedings. Association for Computing Machinery, Inc, pp. 65-72, International Workshop on Innovative Software Development Methodologies and Practices, InnoSWDev 2014, Hong Kong, China, 11/16/14. https://doi.org/10.1145/2666581.2666582
Phannachitta P, Keung J, Monden A, Matsumoto K. Scaling up analogy-based software effort estimation: A Comparison of multiple hadoop implementation schemes. In International Workshop on Innovative Software Development Methodologies and Practices, InnoSWDev 2014 - Proceedings. Association for Computing Machinery, Inc. 2014. p. 65-72 https://doi.org/10.1145/2666581.2666582
Phannachitta, Passakorn ; Keung, Jacky ; Monden, Akito ; Matsumoto, Kenichi. / Scaling up analogy-based software effort estimation : A Comparison of multiple hadoop implementation schemes. International Workshop on Innovative Software Development Methodologies and Practices, InnoSWDev 2014 - Proceedings. Association for Computing Machinery, Inc, 2014. pp. 65-72
@inproceedings{3f8793e318c947af82960d2bdf649938,
title = "Scaling up analogy-based software effort estimation: A Comparison of multiple hadoop implementation schemes",
abstract = "Analogy-based estimation (ABE) is one of the most time consuming and compute intensive method in software de- velopment effort estimation. Optimizing ABE has been a dilemma because simplifying the procedure can reduce the estimation performance, while increasing the procedure com- plexity with more sophisticated theory may sacrifice an ad- vantage of the unlimited scalability for a large data input. Motivated by an emergence of cloud computing technology in software applications, in this study we present 3 different implementation schemes based on Hadoop MapReduce to optimize the ABE process across multiple computing in- stances in the cloud-computing environment. We experimentally compared the 3 MapReduce implementation schemes in contrast with our previously proposed GPGPU approach (named ABE-CUDA) over 8 high-performance Amazon EC2 instances. Results present that the Hadoop solution can pro- vide more computational resources that can extend the scalability of the ABE process. We recommend adoption of 2 different Hadoop implementations (Hadoop streaming and RHadoop) for accelerating the computation specifically for compute-intensive software engineering related tasks.",
keywords = "Analogy-based estimation, Cloud computing, CUDA, Map reduce, Software effort estimation",
author = "Passakorn Phannachitta and Jacky Keung and Akito Monden and Kenichi Matsumoto",
year = "2014",
month = "11",
day = "16",
doi = "10.1145/2666581.2666582",
language = "English",
isbn = "9781450332262",
pages = "65--72",
booktitle = "International Workshop on Innovative Software Development Methodologies and Practices, InnoSWDev 2014 - Proceedings",
publisher = "Association for Computing Machinery, Inc",

}

TY - GEN

T1 - Scaling up analogy-based software effort estimation

T2 - A Comparison of multiple hadoop implementation schemes

AU - Phannachitta, Passakorn

AU - Keung, Jacky

AU - Monden, Akito

AU - Matsumoto, Kenichi

PY - 2014/11/16

Y1 - 2014/11/16

N2 - Analogy-based estimation (ABE) is one of the most time consuming and compute intensive method in software de- velopment effort estimation. Optimizing ABE has been a dilemma because simplifying the procedure can reduce the estimation performance, while increasing the procedure com- plexity with more sophisticated theory may sacrifice an ad- vantage of the unlimited scalability for a large data input. Motivated by an emergence of cloud computing technology in software applications, in this study we present 3 different implementation schemes based on Hadoop MapReduce to optimize the ABE process across multiple computing in- stances in the cloud-computing environment. We experimentally compared the 3 MapReduce implementation schemes in contrast with our previously proposed GPGPU approach (named ABE-CUDA) over 8 high-performance Amazon EC2 instances. Results present that the Hadoop solution can pro- vide more computational resources that can extend the scalability of the ABE process. We recommend adoption of 2 different Hadoop implementations (Hadoop streaming and RHadoop) for accelerating the computation specifically for compute-intensive software engineering related tasks.

AB - Analogy-based estimation (ABE) is one of the most time consuming and compute intensive method in software de- velopment effort estimation. Optimizing ABE has been a dilemma because simplifying the procedure can reduce the estimation performance, while increasing the procedure com- plexity with more sophisticated theory may sacrifice an ad- vantage of the unlimited scalability for a large data input. Motivated by an emergence of cloud computing technology in software applications, in this study we present 3 different implementation schemes based on Hadoop MapReduce to optimize the ABE process across multiple computing in- stances in the cloud-computing environment. We experimentally compared the 3 MapReduce implementation schemes in contrast with our previously proposed GPGPU approach (named ABE-CUDA) over 8 high-performance Amazon EC2 instances. Results present that the Hadoop solution can pro- vide more computational resources that can extend the scalability of the ABE process. We recommend adoption of 2 different Hadoop implementations (Hadoop streaming and RHadoop) for accelerating the computation specifically for compute-intensive software engineering related tasks.

KW - Analogy-based estimation

KW - Cloud computing

KW - CUDA

KW - Map reduce

KW - Software effort estimation

UR - http://www.scopus.com/inward/record.url?scp=84942782262&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84942782262&partnerID=8YFLogxK

U2 - 10.1145/2666581.2666582

DO - 10.1145/2666581.2666582

M3 - Conference contribution

AN - SCOPUS:84942782262

SN - 9781450332262

SP - 65

EP - 72

BT - International Workshop on Innovative Software Development Methodologies and Practices, InnoSWDev 2014 - Proceedings

PB - Association for Computing Machinery, Inc

ER -