A learning-based framework for optimizing service migration in mobile edge clouds

Florian Brandherm, Lin Wang, Max Mühlhäuser

Research output: Chapter in Book / Report / Conference proceedingConference contributionAcademicpeer-review

Abstract

Mobile edge computing is gaining traction due to its ability to deliver ultra-low-latency services for mobile applications. This is achieved through a federation of edge clouds in close proximity of users. However, the intrinsic mobility of users brings a high level of dynamics to the edge environment, calling for sophisticated service migration management across the edge clouds. Previous solutions for edge service placement/migration are architecture-specific, centralized, or are based on restricted cost models. These limitations leave doubts about the practicality of these approaches due to the lack of a standardized reference model for edge clouds. In this paper, we propose a general framework for optimizing edge service migration based on reinforcement learning techniques. Using our framework, edge service migration strategies can be learned with respect to a large variety of optimization goals. Moreover, our learning-based algorithm is agnostic to the underlying architecture and resource constraints. Preliminary results show that our model-free learning-based approach can compete with model-based baselines and adapt to different objectives.

Original languageEnglish
Title of host publicationEdgeSys 2019 - Proceedings of the 2nd ACM International Workshop on Edge Systems, Analytics and Networking, Part of EuroSys 2019
PublisherAssociation for Computing Machinery, Inc
Pages12-17
Number of pages6
ISBN (Electronic)9781450362757
DOIs
Publication statusPublished - 25 Mar 2019
Event2nd ACM International Workshop on Edge Systems, Analytics and Networking, EdgeSys 2019, Part of EuroSys 2019 - Dresden, Germany
Duration: 25 Mar 201925 Mar 2019

Conference

Conference2nd ACM International Workshop on Edge Systems, Analytics and Networking, EdgeSys 2019, Part of EuroSys 2019
CountryGermany
CityDresden
Period25/03/1925/03/19

Fingerprint

Reinforcement learning
Costs

Cite this

Brandherm, F., Wang, L., & Mühlhäuser, M. (2019). A learning-based framework for optimizing service migration in mobile edge clouds. In EdgeSys 2019 - Proceedings of the 2nd ACM International Workshop on Edge Systems, Analytics and Networking, Part of EuroSys 2019 (pp. 12-17). Association for Computing Machinery, Inc. https://doi.org/10.1145/3301418.3313939
Brandherm, Florian ; Wang, Lin ; Mühlhäuser, Max. / A learning-based framework for optimizing service migration in mobile edge clouds. EdgeSys 2019 - Proceedings of the 2nd ACM International Workshop on Edge Systems, Analytics and Networking, Part of EuroSys 2019. Association for Computing Machinery, Inc, 2019. pp. 12-17
@inproceedings{27895c4e154a44b6bcbce9331f3b3b83,
title = "A learning-based framework for optimizing service migration in mobile edge clouds",
abstract = "Mobile edge computing is gaining traction due to its ability to deliver ultra-low-latency services for mobile applications. This is achieved through a federation of edge clouds in close proximity of users. However, the intrinsic mobility of users brings a high level of dynamics to the edge environment, calling for sophisticated service migration management across the edge clouds. Previous solutions for edge service placement/migration are architecture-specific, centralized, or are based on restricted cost models. These limitations leave doubts about the practicality of these approaches due to the lack of a standardized reference model for edge clouds. In this paper, we propose a general framework for optimizing edge service migration based on reinforcement learning techniques. Using our framework, edge service migration strategies can be learned with respect to a large variety of optimization goals. Moreover, our learning-based algorithm is agnostic to the underlying architecture and resource constraints. Preliminary results show that our model-free learning-based approach can compete with model-based baselines and adapt to different objectives.",
author = "Florian Brandherm and Lin Wang and Max M{\"u}hlh{\"a}user",
year = "2019",
month = "3",
day = "25",
doi = "10.1145/3301418.3313939",
language = "English",
pages = "12--17",
booktitle = "EdgeSys 2019 - Proceedings of the 2nd ACM International Workshop on Edge Systems, Analytics and Networking, Part of EuroSys 2019",
publisher = "Association for Computing Machinery, Inc",

}

Brandherm, F, Wang, L & Mühlhäuser, M 2019, A learning-based framework for optimizing service migration in mobile edge clouds. in EdgeSys 2019 - Proceedings of the 2nd ACM International Workshop on Edge Systems, Analytics and Networking, Part of EuroSys 2019. Association for Computing Machinery, Inc, pp. 12-17, 2nd ACM International Workshop on Edge Systems, Analytics and Networking, EdgeSys 2019, Part of EuroSys 2019, Dresden, Germany, 25/03/19. https://doi.org/10.1145/3301418.3313939

A learning-based framework for optimizing service migration in mobile edge clouds. / Brandherm, Florian; Wang, Lin; Mühlhäuser, Max.

EdgeSys 2019 - Proceedings of the 2nd ACM International Workshop on Edge Systems, Analytics and Networking, Part of EuroSys 2019. Association for Computing Machinery, Inc, 2019. p. 12-17.

Research output: Chapter in Book / Report / Conference proceedingConference contributionAcademicpeer-review

TY - GEN

T1 - A learning-based framework for optimizing service migration in mobile edge clouds

AU - Brandherm, Florian

AU - Wang, Lin

AU - Mühlhäuser, Max

PY - 2019/3/25

Y1 - 2019/3/25

N2 - Mobile edge computing is gaining traction due to its ability to deliver ultra-low-latency services for mobile applications. This is achieved through a federation of edge clouds in close proximity of users. However, the intrinsic mobility of users brings a high level of dynamics to the edge environment, calling for sophisticated service migration management across the edge clouds. Previous solutions for edge service placement/migration are architecture-specific, centralized, or are based on restricted cost models. These limitations leave doubts about the practicality of these approaches due to the lack of a standardized reference model for edge clouds. In this paper, we propose a general framework for optimizing edge service migration based on reinforcement learning techniques. Using our framework, edge service migration strategies can be learned with respect to a large variety of optimization goals. Moreover, our learning-based algorithm is agnostic to the underlying architecture and resource constraints. Preliminary results show that our model-free learning-based approach can compete with model-based baselines and adapt to different objectives.

AB - Mobile edge computing is gaining traction due to its ability to deliver ultra-low-latency services for mobile applications. This is achieved through a federation of edge clouds in close proximity of users. However, the intrinsic mobility of users brings a high level of dynamics to the edge environment, calling for sophisticated service migration management across the edge clouds. Previous solutions for edge service placement/migration are architecture-specific, centralized, or are based on restricted cost models. These limitations leave doubts about the practicality of these approaches due to the lack of a standardized reference model for edge clouds. In this paper, we propose a general framework for optimizing edge service migration based on reinforcement learning techniques. Using our framework, edge service migration strategies can be learned with respect to a large variety of optimization goals. Moreover, our learning-based algorithm is agnostic to the underlying architecture and resource constraints. Preliminary results show that our model-free learning-based approach can compete with model-based baselines and adapt to different objectives.

UR - http://www.scopus.com/inward/record.url?scp=85063874533&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85063874533&partnerID=8YFLogxK

U2 - 10.1145/3301418.3313939

DO - 10.1145/3301418.3313939

M3 - Conference contribution

SP - 12

EP - 17

BT - EdgeSys 2019 - Proceedings of the 2nd ACM International Workshop on Edge Systems, Analytics and Networking, Part of EuroSys 2019

PB - Association for Computing Machinery, Inc

ER -

Brandherm F, Wang L, Mühlhäuser M. A learning-based framework for optimizing service migration in mobile edge clouds. In EdgeSys 2019 - Proceedings of the 2nd ACM International Workshop on Edge Systems, Analytics and Networking, Part of EuroSys 2019. Association for Computing Machinery, Inc. 2019. p. 12-17 https://doi.org/10.1145/3301418.3313939