BigDataBench-MT: A benchmark tool for generating realistic mixed data center workloads

Rui Han; Shulin Zhan; Chenrong Shao; Junwei Wang; Lizy K. John; Jiangtao Xu; Gang Lu; Lei Wang

doi:10.1007/978-3-319-29006-5_2

BigDataBench-MT: A benchmark tool for generating realistic mixed data center workloads

Rui Han^*
, Shulin Zhan
, Chenrong Shao
, Junwei Wang
, Lizy K. John
, Jiangtao Xu
, Gang Lu
, Lei Wang

^*Corresponding author for this work

Chinese Academy of Sciences
ICarsclub
Xi'an Jiaotong University
Kingsoft Cloud
University of Texas at Austin
University of Chinese Academy of Sciences

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

4 Citations (Scopus)

Abstract

Long-running service workloads (e.g. web search engine) and short-term data analysis workloads (e.g. Hadoop MapReduce jobs) colocate in today’s data centers. Developing realistic benchmarks to reflect such practical scenario of mixed workload is a key problem to produce trustworthy results when evaluating and comparing data center systems. This requires using actual workloads as well as guaranteeing their submissions to follow patterns hidden in real-world traces. However, existing benchmarks either generate actual workloads based on probability models, or replay real-world workload traces using basic I/O operations. To fill this gap, we propose a benchmark tool that is a first step towards generating a mix of actual service and data analysis workloads on the basis of real workload traces. Our tool includes a combiner that enables the replaying of actual workloads according to the workload traces, and a multi-tenant generator that flexibly scales the workloads up and down according to users’ requirements. Based on this, our demo illustrates the workload customization and generation process using a visual interface. The proposed tool, called BigDataBench-MT, is a multitenant version of our comprehensive benchmark suite BigDataBench and it is publicly available from http://prof.ict.ac.cn/BigDataBench/ multi-tenancyversion/.

Original language	English
Title of host publication	Big Data Benchmarks, Performance Optimization, and Emerging Hardware - 6th Workshop, BPOE 2015, Revised Selected Papers
Editors	Roberto V. Zicari, Jianfeng Zhan, Rui Han
Publisher	Springer Verlag
Pages	10-21
Number of pages	12
ISBN (Print)	9783319290058
DOIs	https://doi.org/10.1007/978-3-319-29006-5_2
Publication status	Published - 2016
Externally published	Yes
Event	6th Workshop on Big Data Benchmarks, Performance Optimization, and Emerging Hardware, BPOE 2015 - Kohala, United States Duration: 31 Aug 2015 → 4 Sept 2015

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	9495
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	6th Workshop on Big Data Benchmarks, Performance Optimization, and Emerging Hardware, BPOE 2015
Country/Territory	United States
City	Kohala
Period	31/08/15 → 4/09/15

Keywords

Benchmark
Data center
Mixed workloads
Workload trace

Access to Document

10.1007/978-3-319-29006-5_2

Cite this

Han, R., Zhan, S., Shao, C., Wang, J., John, L. K., Xu, J., Lu, G., & Wang, L. (2016). BigDataBench-MT: A benchmark tool for generating realistic mixed data center workloads. In R. V. Zicari, J. Zhan, & R. Han (Eds.), Big Data Benchmarks, Performance Optimization, and Emerging Hardware - 6th Workshop, BPOE 2015, Revised Selected Papers (pp. 10-21). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 9495). Springer Verlag. https://doi.org/10.1007/978-3-319-29006-5_2

Han, Rui ; Zhan, Shulin ; Shao, Chenrong et al. / BigDataBench-MT : A benchmark tool for generating realistic mixed data center workloads. Big Data Benchmarks, Performance Optimization, and Emerging Hardware - 6th Workshop, BPOE 2015, Revised Selected Papers. editor / Roberto V. Zicari ; Jianfeng Zhan ; Rui Han. Springer Verlag, 2016. pp. 10-21 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{82f1e016506d49e1b3e30a338be752ab,

title = "BigDataBench-MT: A benchmark tool for generating realistic mixed data center workloads",

abstract = "Long-running service workloads (e.g. web search engine) and short-term data analysis workloads (e.g. Hadoop MapReduce jobs) colocate in today{\textquoteright}s data centers. Developing realistic benchmarks to reflect such practical scenario of mixed workload is a key problem to produce trustworthy results when evaluating and comparing data center systems. This requires using actual workloads as well as guaranteeing their submissions to follow patterns hidden in real-world traces. However, existing benchmarks either generate actual workloads based on probability models, or replay real-world workload traces using basic I/O operations. To fill this gap, we propose a benchmark tool that is a first step towards generating a mix of actual service and data analysis workloads on the basis of real workload traces. Our tool includes a combiner that enables the replaying of actual workloads according to the workload traces, and a multi-tenant generator that flexibly scales the workloads up and down according to users{\textquoteright} requirements. Based on this, our demo illustrates the workload customization and generation process using a visual interface. The proposed tool, called BigDataBench-MT, is a multitenant version of our comprehensive benchmark suite BigDataBench and it is publicly available from http://prof.ict.ac.cn/BigDataBench/ multi-tenancyversion/.",

keywords = "Benchmark, Data center, Mixed workloads, Workload trace",

author = "Rui Han and Shulin Zhan and Chenrong Shao and Junwei Wang and John, \{Lizy K.\} and Jiangtao Xu and Gang Lu and Lei Wang",

note = "Publisher Copyright: {\textcopyright} Springer International Publishing Switzerland 2016.; 6th Workshop on Big Data Benchmarks, Performance Optimization, and Emerging Hardware, BPOE 2015 ; Conference date: 31-08-2015 Through 04-09-2015",

year = "2016",

doi = "10.1007/978-3-319-29006-5\_2",

language = "English",

isbn = "9783319290058",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

publisher = "Springer Verlag",

pages = "10--21",

editor = "Zicari, \{Roberto V.\} and Jianfeng Zhan and Rui Han",

booktitle = "Big Data Benchmarks, Performance Optimization, and Emerging Hardware - 6th Workshop, BPOE 2015, Revised Selected Papers",

address = "Germany",

}

Han, R, Zhan, S, Shao, C, Wang, J, John, LK, Xu, J, Lu, G & Wang, L 2016, BigDataBench-MT: A benchmark tool for generating realistic mixed data center workloads. in RV Zicari, J Zhan & R Han (eds), Big Data Benchmarks, Performance Optimization, and Emerging Hardware - 6th Workshop, BPOE 2015, Revised Selected Papers. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 9495, Springer Verlag, pp. 10-21, 6th Workshop on Big Data Benchmarks, Performance Optimization, and Emerging Hardware, BPOE 2015, Kohala, United States, 31/08/15. https://doi.org/10.1007/978-3-319-29006-5_2

BigDataBench-MT: A benchmark tool for generating realistic mixed data center workloads. / Han, Rui; Zhan, Shulin; Shao, Chenrong et al.
Big Data Benchmarks, Performance Optimization, and Emerging Hardware - 6th Workshop, BPOE 2015, Revised Selected Papers. ed. / Roberto V. Zicari; Jianfeng Zhan; Rui Han. Springer Verlag, 2016. p. 10-21 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 9495).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - BigDataBench-MT

T2 - 6th Workshop on Big Data Benchmarks, Performance Optimization, and Emerging Hardware, BPOE 2015

AU - Han, Rui

AU - Zhan, Shulin

AU - Shao, Chenrong

AU - Wang, Junwei

AU - John, Lizy K.

AU - Xu, Jiangtao

AU - Lu, Gang

AU - Wang, Lei

N1 - Publisher Copyright: © Springer International Publishing Switzerland 2016.

PY - 2016

Y1 - 2016

N2 - Long-running service workloads (e.g. web search engine) and short-term data analysis workloads (e.g. Hadoop MapReduce jobs) colocate in today’s data centers. Developing realistic benchmarks to reflect such practical scenario of mixed workload is a key problem to produce trustworthy results when evaluating and comparing data center systems. This requires using actual workloads as well as guaranteeing their submissions to follow patterns hidden in real-world traces. However, existing benchmarks either generate actual workloads based on probability models, or replay real-world workload traces using basic I/O operations. To fill this gap, we propose a benchmark tool that is a first step towards generating a mix of actual service and data analysis workloads on the basis of real workload traces. Our tool includes a combiner that enables the replaying of actual workloads according to the workload traces, and a multi-tenant generator that flexibly scales the workloads up and down according to users’ requirements. Based on this, our demo illustrates the workload customization and generation process using a visual interface. The proposed tool, called BigDataBench-MT, is a multitenant version of our comprehensive benchmark suite BigDataBench and it is publicly available from http://prof.ict.ac.cn/BigDataBench/ multi-tenancyversion/.

AB - Long-running service workloads (e.g. web search engine) and short-term data analysis workloads (e.g. Hadoop MapReduce jobs) colocate in today’s data centers. Developing realistic benchmarks to reflect such practical scenario of mixed workload is a key problem to produce trustworthy results when evaluating and comparing data center systems. This requires using actual workloads as well as guaranteeing their submissions to follow patterns hidden in real-world traces. However, existing benchmarks either generate actual workloads based on probability models, or replay real-world workload traces using basic I/O operations. To fill this gap, we propose a benchmark tool that is a first step towards generating a mix of actual service and data analysis workloads on the basis of real workload traces. Our tool includes a combiner that enables the replaying of actual workloads according to the workload traces, and a multi-tenant generator that flexibly scales the workloads up and down according to users’ requirements. Based on this, our demo illustrates the workload customization and generation process using a visual interface. The proposed tool, called BigDataBench-MT, is a multitenant version of our comprehensive benchmark suite BigDataBench and it is publicly available from http://prof.ict.ac.cn/BigDataBench/ multi-tenancyversion/.

KW - Benchmark

KW - Data center

KW - Mixed workloads

KW - Workload trace

UR - https://www.scopus.com/pages/publications/84958057085

U2 - 10.1007/978-3-319-29006-5_2

DO - 10.1007/978-3-319-29006-5_2

M3 - Conference contribution

AN - SCOPUS:84958057085

SN - 9783319290058

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 10

EP - 21

BT - Big Data Benchmarks, Performance Optimization, and Emerging Hardware - 6th Workshop, BPOE 2015, Revised Selected Papers

A2 - Zicari, Roberto V.

A2 - Zhan, Jianfeng

A2 - Han, Rui

PB - Springer Verlag

Y2 - 31 August 2015 through 4 September 2015

ER -

Han R, Zhan S, Shao C, Wang J, John LK, Xu J et al. BigDataBench-MT: A benchmark tool for generating realistic mixed data center workloads. In Zicari RV, Zhan J, Han R, editors, Big Data Benchmarks, Performance Optimization, and Emerging Hardware - 6th Workshop, BPOE 2015, Revised Selected Papers. Springer Verlag. 2016. p. 10-21. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). doi: 10.1007/978-3-319-29006-5_2

BigDataBench-MT: A benchmark tool for generating realistic mixed data center workloads

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this