Putting Data Science Pipelines on the Edge

Ali Akoglu; Genoveva Vargas-Solar

doi:10.1007/978-3-030-92231-3_1

Putting Data Science Pipelines on the Edge

Ali Akoglu, Genoveva Vargas-Solar

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

Abstract

This paper proposes a composable “Just in Time Architecture” for Data Science (DS) Pipelines named JITA-4DS and associated resource management techniques for configuring disaggregated data centers (DCs). DCs under our approach are composable based on vertical integration of the application, middleware/operating system, and hardware layers customized dynamically to meet application Service Level Objectives (SLO - application-aware management). Thereby, pipelines utilize a set of flexible building blocks that can be dynamically and automatically assembled and re-assembled to meet the dynamic changes in the workload’s SLOs. To assess disaggregated DC’s, we study how to model and validate their performance in large-scale settings.

Original language	English (US)
Title of host publication	ICWE 2021 Workshops - ICWE 2021 International Workshops, BECS and Invited Papers, 2021, Revised Selected Papers
Editors	Maxim Bakaev, In-Young Ko, Michael Mrissa, Michael Mrissa, Cesare Pautasso, Abhishek Srivastava
Publisher	Springer Science and Business Media Deutschland GmbH
Pages	3-14
Number of pages	12
ISBN (Print)	9783030922306
DOIs	https://doi.org/10.1007/978-3-030-92231-3_1
State	Published - 2022
Event	1st International Workshop on Big data-driven Edge Cloud Services, BECS 2021 held in conjunction with 21st International Conference on Web Engineering, ICWE 2021 - Virtual, Online Duration: May 18 2021 → May 21 2021

Publication series

Name	Communications in Computer and Information Science
Volume	1508 CCIS
ISSN (Print)	1865-0929
ISSN (Electronic)	1865-0937

Conference

Conference	1st International Workshop on Big data-driven Edge Cloud Services, BECS 2021 held in conjunction with 21st International Conference on Web Engineering, ICWE 2021
City	Virtual, Online
Period	5/18/21 → 5/21/21

Keywords

Data science pipelines
Disaggregated data centers
Edge computing

ASJC Scopus subject areas

General Computer Science
General Mathematics

Access to Document

10.1007/978-3-030-92231-3_1

Cite this

Akoglu, A., & Vargas-Solar, G. (2022). Putting Data Science Pipelines on the Edge. In M. Bakaev, I.-Y. Ko, M. Mrissa, M. Mrissa, C. Pautasso, & A. Srivastava (Eds.), ICWE 2021 Workshops - ICWE 2021 International Workshops, BECS and Invited Papers, 2021, Revised Selected Papers (pp. 3-14). (Communications in Computer and Information Science; Vol. 1508 CCIS). Springer Science and Business Media Deutschland GmbH. https://doi.org/10.1007/978-3-030-92231-3_1

Putting Data Science Pipelines on the Edge. / Akoglu, Ali; Vargas-Solar, Genoveva.
ICWE 2021 Workshops - ICWE 2021 International Workshops, BECS and Invited Papers, 2021, Revised Selected Papers. ed. / Maxim Bakaev; In-Young Ko; Michael Mrissa; Michael Mrissa; Cesare Pautasso; Abhishek Srivastava. Springer Science and Business Media Deutschland GmbH, 2022. p. 3-14 (Communications in Computer and Information Science; Vol. 1508 CCIS).

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

Akoglu, A & Vargas-Solar, G 2022, Putting Data Science Pipelines on the Edge. in M Bakaev, I-Y Ko, M Mrissa, M Mrissa, C Pautasso & A Srivastava (eds), ICWE 2021 Workshops - ICWE 2021 International Workshops, BECS and Invited Papers, 2021, Revised Selected Papers. Communications in Computer and Information Science, vol. 1508 CCIS, Springer Science and Business Media Deutschland GmbH, pp. 3-14, 1st International Workshop on Big data-driven Edge Cloud Services, BECS 2021 held in conjunction with 21st International Conference on Web Engineering, ICWE 2021, Virtual, Online, 5/18/21. https://doi.org/10.1007/978-3-030-92231-3_1

Akoglu A, Vargas-Solar G. Putting Data Science Pipelines on the Edge. In Bakaev M, Ko IY, Mrissa M, Mrissa M, Pautasso C, Srivastava A, editors, ICWE 2021 Workshops - ICWE 2021 International Workshops, BECS and Invited Papers, 2021, Revised Selected Papers. Springer Science and Business Media Deutschland GmbH. 2022. p. 3-14. (Communications in Computer and Information Science). doi: 10.1007/978-3-030-92231-3_1

Akoglu, Ali ; Vargas-Solar, Genoveva. / Putting Data Science Pipelines on the Edge. ICWE 2021 Workshops - ICWE 2021 International Workshops, BECS and Invited Papers, 2021, Revised Selected Papers. editor / Maxim Bakaev ; In-Young Ko ; Michael Mrissa ; Michael Mrissa ; Cesare Pautasso ; Abhishek Srivastava. Springer Science and Business Media Deutschland GmbH, 2022. pp. 3-14 (Communications in Computer and Information Science).

@inproceedings{ebcf13a49fbb42b6b6fda87df0416336,

title = "Putting Data Science Pipelines on the Edge",

abstract = "This paper proposes a composable “Just in Time Architecture” for Data Science (DS) Pipelines named JITA-4DS and associated resource management techniques for configuring disaggregated data centers (DCs). DCs under our approach are composable based on vertical integration of the application, middleware/operating system, and hardware layers customized dynamically to meet application Service Level Objectives (SLO - application-aware management). Thereby, pipelines utilize a set of flexible building blocks that can be dynamically and automatically assembled and re-assembled to meet the dynamic changes in the workload{\textquoteright}s SLOs. To assess disaggregated DC{\textquoteright}s, we study how to model and validate their performance in large-scale settings.",

keywords = "Data science pipelines, Disaggregated data centers, Edge computing",

author = "Ali Akoglu and Genoveva Vargas-Solar",

note = "Publisher Copyright: {\textcopyright} 2022, Springer Nature Switzerland AG.; 1st International Workshop on Big data-driven Edge Cloud Services, BECS 2021 held in conjunction with 21st International Conference on Web Engineering, ICWE 2021 ; Conference date: 18-05-2021 Through 21-05-2021",

year = "2022",

doi = "10.1007/978-3-030-92231-3_1",

language = "English (US)",

isbn = "9783030922306",

series = "Communications in Computer and Information Science",

publisher = "Springer Science and Business Media Deutschland GmbH",

pages = "3--14",

editor = "Maxim Bakaev and In-Young Ko and Michael Mrissa and Michael Mrissa and Cesare Pautasso and Abhishek Srivastava",

booktitle = "ICWE 2021 Workshops - ICWE 2021 International Workshops, BECS and Invited Papers, 2021, Revised Selected Papers",

address = "Germany",

}

TY - GEN

T1 - Putting Data Science Pipelines on the Edge

AU - Akoglu, Ali

AU - Vargas-Solar, Genoveva

PY - 2022

Y1 - 2022

N2 - This paper proposes a composable “Just in Time Architecture” for Data Science (DS) Pipelines named JITA-4DS and associated resource management techniques for configuring disaggregated data centers (DCs). DCs under our approach are composable based on vertical integration of the application, middleware/operating system, and hardware layers customized dynamically to meet application Service Level Objectives (SLO - application-aware management). Thereby, pipelines utilize a set of flexible building blocks that can be dynamically and automatically assembled and re-assembled to meet the dynamic changes in the workload’s SLOs. To assess disaggregated DC’s, we study how to model and validate their performance in large-scale settings.

AB - This paper proposes a composable “Just in Time Architecture” for Data Science (DS) Pipelines named JITA-4DS and associated resource management techniques for configuring disaggregated data centers (DCs). DCs under our approach are composable based on vertical integration of the application, middleware/operating system, and hardware layers customized dynamically to meet application Service Level Objectives (SLO - application-aware management). Thereby, pipelines utilize a set of flexible building blocks that can be dynamically and automatically assembled and re-assembled to meet the dynamic changes in the workload’s SLOs. To assess disaggregated DC’s, we study how to model and validate their performance in large-scale settings.

KW - Data science pipelines

KW - Disaggregated data centers

KW - Edge computing

UR - http://www.scopus.com/inward/record.url?scp=85121916758&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85121916758&partnerID=8YFLogxK

U2 - 10.1007/978-3-030-92231-3_1

DO - 10.1007/978-3-030-92231-3_1

M3 - Conference contribution

AN - SCOPUS:85121916758

SN - 9783030922306

T3 - Communications in Computer and Information Science

SP - 3

EP - 14

BT - ICWE 2021 Workshops - ICWE 2021 International Workshops, BECS and Invited Papers, 2021, Revised Selected Papers

A2 - Bakaev, Maxim

A2 - Ko, In-Young

A2 - Mrissa, Michael

A2 - Pautasso, Cesare

A2 - Srivastava, Abhishek

PB - Springer Science and Business Media Deutschland GmbH

T2 - 1st International Workshop on Big data-driven Edge Cloud Services, BECS 2021 held in conjunction with 21st International Conference on Web Engineering, ICWE 2021

Y2 - 18 May 2021 through 21 May 2021

ER -