Fitting realistic data centre workloads a data science approach

Björn F. Postema; Niels J. Geuze; Boudewijn R. Haverkort

doi:10.1145/3208903.3213520

Fitting realistic data centre workloads a data science approach

Björn F. Postema
, Niels J. Geuze
, Boudewijn R. Haverkort

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

2 Citations (Scopus)

Abstract

Data centres are playing a pivotal role in all cloud-based services (e-commerce, social networks, financial services, e-government, etc.). The performance of data centres is crucial for the acceptance of all these services by end-users. It is important to carefully design data centres with both performance and energy considerations in mind, as data centres are also known to use large amounts of electrical energy. For that purpose we have developed a modular simulation model (based on Anylogic) that can be used to study performance-energy trade-offs in data centre design. Key to such studies is the availability of a workload model. In this paper we present a workload characterisation model and algorithm using modern-day data science techniques, building on top of Jupyter Notebook and the ProFiDo platform. We present the method and show its versatility on a case study with real-world traces of 20 million entries, provided by the Dutch company better.be.

Original language	English
Title of host publication	e-Energy 2018 - Proceedings of the 9th ACM International Conference on Future Energy Systems
Publisher	Association for Computing Machinery
Pages	486-491
Number of pages	6
ISBN (Electronic)	9781450357678
DOIs	https://doi.org/10.1145/3208903.3213520
Publication status	Published - 12 Jun 2018
Externally published	Yes
Event	9th ACM International Conference on Future Energy Systems, e-Energy 2018 - Karlsruhe, Germany Duration: 12 Jun 2018 → 15 Jun 2018

Publication series

Name	e-Energy 2018 - Proceedings of the 9th ACM International Conference on Future Energy Systems

Conference

Conference	9th ACM International Conference on Future Energy Systems, e-Energy 2018
Country/Territory	Germany
City	Karlsruhe
Period	12/06/18 → 15/06/18

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

SDG 16 Peace, Justice and Strong Institutions

Keywords

Data centre
Data science
Distribution fitting
Mixture normal distribution
Workload modelling

Access to Document

10.1145/3208903.3213520

Cite this

Postema, B. F., Geuze, N. J., & Haverkort, B. R. (2018). Fitting realistic data centre workloads a data science approach. In e-Energy 2018 - Proceedings of the 9th ACM International Conference on Future Energy Systems (pp. 486-491). (e-Energy 2018 - Proceedings of the 9th ACM International Conference on Future Energy Systems). Association for Computing Machinery. https://doi.org/10.1145/3208903.3213520

Postema, Björn F. ; Geuze, Niels J. ; Haverkort, Boudewijn R. / Fitting realistic data centre workloads a data science approach. e-Energy 2018 - Proceedings of the 9th ACM International Conference on Future Energy Systems. Association for Computing Machinery, 2018. pp. 486-491 (e-Energy 2018 - Proceedings of the 9th ACM International Conference on Future Energy Systems).

@inproceedings{a8d21e547b2a4c619984518631c4da1c,

title = "Fitting realistic data centre workloads a data science approach",

abstract = "Data centres are playing a pivotal role in all cloud-based services (e-commerce, social networks, financial services, e-government, etc.). The performance of data centres is crucial for the acceptance of all these services by end-users. It is important to carefully design data centres with both performance and energy considerations in mind, as data centres are also known to use large amounts of electrical energy. For that purpose we have developed a modular simulation model (based on Anylogic) that can be used to study performance-energy trade-offs in data centre design. Key to such studies is the availability of a workload model. In this paper we present a workload characterisation model and algorithm using modern-day data science techniques, building on top of Jupyter Notebook and the ProFiDo platform. We present the method and show its versatility on a case study with real-world traces of 20 million entries, provided by the Dutch company better.be.",

keywords = "Data centre, Data science, Distribution fitting, Mixture normal distribution, Workload modelling",

author = "Postema, \{Bj{\"o}rn F.\} and Geuze, \{Niels J.\} and Haverkort, \{Boudewijn R.\}",

note = "Publisher Copyright: {\textcopyright} 2018 Copyright held by the owner/author(s). Copyright: Copyright 2018 Elsevier B.V., All rights reserved.; 9th ACM International Conference on Future Energy Systems, e-Energy 2018 ; Conference date: 12-06-2018 Through 15-06-2018",

year = "2018",

month = jun,

day = "12",

doi = "10.1145/3208903.3213520",

language = "English",

series = "e-Energy 2018 - Proceedings of the 9th ACM International Conference on Future Energy Systems",

publisher = "Association for Computing Machinery",

pages = "486--491",

booktitle = "e-Energy 2018 - Proceedings of the 9th ACM International Conference on Future Energy Systems",

address = "United States",

}

Postema, BF, Geuze, NJ & Haverkort, BR 2018, Fitting realistic data centre workloads a data science approach. in e-Energy 2018 - Proceedings of the 9th ACM International Conference on Future Energy Systems. e-Energy 2018 - Proceedings of the 9th ACM International Conference on Future Energy Systems, Association for Computing Machinery, pp. 486-491, 9th ACM International Conference on Future Energy Systems, e-Energy 2018, Karlsruhe, Germany, 12/06/18. https://doi.org/10.1145/3208903.3213520

Fitting realistic data centre workloads a data science approach. / Postema, Björn F.; Geuze, Niels J.; Haverkort, Boudewijn R.
e-Energy 2018 - Proceedings of the 9th ACM International Conference on Future Energy Systems. Association for Computing Machinery, 2018. p. 486-491 (e-Energy 2018 - Proceedings of the 9th ACM International Conference on Future Energy Systems).

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

TY - GEN

T1 - Fitting realistic data centre workloads a data science approach

AU - Postema, Björn F.

AU - Geuze, Niels J.

AU - Haverkort, Boudewijn R.

PY - 2018/6/12

Y1 - 2018/6/12

N2 - Data centres are playing a pivotal role in all cloud-based services (e-commerce, social networks, financial services, e-government, etc.). The performance of data centres is crucial for the acceptance of all these services by end-users. It is important to carefully design data centres with both performance and energy considerations in mind, as data centres are also known to use large amounts of electrical energy. For that purpose we have developed a modular simulation model (based on Anylogic) that can be used to study performance-energy trade-offs in data centre design. Key to such studies is the availability of a workload model. In this paper we present a workload characterisation model and algorithm using modern-day data science techniques, building on top of Jupyter Notebook and the ProFiDo platform. We present the method and show its versatility on a case study with real-world traces of 20 million entries, provided by the Dutch company better.be.

AB - Data centres are playing a pivotal role in all cloud-based services (e-commerce, social networks, financial services, e-government, etc.). The performance of data centres is crucial for the acceptance of all these services by end-users. It is important to carefully design data centres with both performance and energy considerations in mind, as data centres are also known to use large amounts of electrical energy. For that purpose we have developed a modular simulation model (based on Anylogic) that can be used to study performance-energy trade-offs in data centre design. Key to such studies is the availability of a workload model. In this paper we present a workload characterisation model and algorithm using modern-day data science techniques, building on top of Jupyter Notebook and the ProFiDo platform. We present the method and show its versatility on a case study with real-world traces of 20 million entries, provided by the Dutch company better.be.

KW - Data centre

KW - Data science

KW - Distribution fitting

KW - Mixture normal distribution

KW - Workload modelling

UR - https://www.scopus.com/pages/publications/85050194379

U2 - 10.1145/3208903.3213520

DO - 10.1145/3208903.3213520

M3 - Conference contribution

AN - SCOPUS:85050194379

T3 - e-Energy 2018 - Proceedings of the 9th ACM International Conference on Future Energy Systems

SP - 486

EP - 491

BT - e-Energy 2018 - Proceedings of the 9th ACM International Conference on Future Energy Systems

PB - Association for Computing Machinery

T2 - 9th ACM International Conference on Future Energy Systems, e-Energy 2018

Y2 - 12 June 2018 through 15 June 2018

ER -

Postema BF, Geuze NJ, Haverkort BR. Fitting realistic data centre workloads a data science approach. In e-Energy 2018 - Proceedings of the 9th ACM International Conference on Future Energy Systems. Association for Computing Machinery. 2018. p. 486-491. (e-Energy 2018 - Proceedings of the 9th ACM International Conference on Future Energy Systems). doi: 10.1145/3208903.3213520

Fitting realistic data centre workloads a data science approach

Abstract

Publication series

Conference

UN SDGs

Keywords

Access to Document

Other files and links

Fingerprint

Cite this