Osmotic computing-based service migration and resource scheduling in Mobile Augmented Reality Networks (MARN)

Vishal Sharma, Dushantha Nalin K. Jayakody, Marwa Qaraqe

Research output: Contribution to journalArticle

Abstract

Resources and services between the servers in Mobile Augmented Reality Networks (MARN) are tedious to manage. These networks comprise users possessing Augmented Reality (AR)-Virtual Reality (VR) applications. Low latency, robustness, and tolerance are the key requirements of these networks, which can be attained by using near-user solutions such as edge computing. However, management of services and scheduling them to near-user servers in an integrated environment of edge and public/private infrastructure are complex tasks. These require an optimal solution, which can be obtained by using “Osmotic Computing”, that has been recently proposed as a paradigm for the integration of edge and public/private cloud. This paper uses osmotic computing for effectively migrating and scheduling the services between the servers of the different layers. The paper also presents the details on various components that are used for applying osmotic computing to a network followed by core applications, types, service classification, migration, and scheduling through the rules of osmotic game formulated for its operations. The evaluations are conducted on 100,000 requests and the proposed approach shows significant performance with the probability of the error being 0.1 at 55.72% conservation of the energy and memory resources for the entire network despite the increasing number of users. The proposed approach also satisfies the conditions of the joint optimization functions presented in the system model and demonstrates that the system holds true even with varying users, thus, proving its robustness and tolerance against the number of users.

Original languageEnglish
Pages (from-to)723-737
Number of pages15
JournalFuture Generation Computer Systems
Volume102
DOIs
Publication statusPublished - Jan 2020

Fingerprint

Augmented reality
Servers
Scheduling
Virtual reality
Conservation
Data storage equipment

Keywords

  • Augmented reality
  • Edge-computing
  • Osmotic computing
  • Resource scheduling
  • Service migrations

ASJC Scopus subject areas

  • Software
  • Hardware and Architecture
  • Computer Networks and Communications

Cite this

Osmotic computing-based service migration and resource scheduling in Mobile Augmented Reality Networks (MARN). / Sharma, Vishal; Jayakody, Dushantha Nalin K.; Qaraqe, Marwa.

In: Future Generation Computer Systems, Vol. 102, 01.2020, p. 723-737.

Research output: Contribution to journalArticle

@article{2eb5c153ee4b4fca84b0bfa382a507d3,
title = "Osmotic computing-based service migration and resource scheduling in Mobile Augmented Reality Networks (MARN)",
abstract = "Resources and services between the servers in Mobile Augmented Reality Networks (MARN) are tedious to manage. These networks comprise users possessing Augmented Reality (AR)-Virtual Reality (VR) applications. Low latency, robustness, and tolerance are the key requirements of these networks, which can be attained by using near-user solutions such as edge computing. However, management of services and scheduling them to near-user servers in an integrated environment of edge and public/private infrastructure are complex tasks. These require an optimal solution, which can be obtained by using “Osmotic Computing”, that has been recently proposed as a paradigm for the integration of edge and public/private cloud. This paper uses osmotic computing for effectively migrating and scheduling the services between the servers of the different layers. The paper also presents the details on various components that are used for applying osmotic computing to a network followed by core applications, types, service classification, migration, and scheduling through the rules of osmotic game formulated for its operations. The evaluations are conducted on 100,000 requests and the proposed approach shows significant performance with the probability of the error being 0.1 at 55.72{\%} conservation of the energy and memory resources for the entire network despite the increasing number of users. The proposed approach also satisfies the conditions of the joint optimization functions presented in the system model and demonstrates that the system holds true even with varying users, thus, proving its robustness and tolerance against the number of users.",
keywords = "Augmented reality, Edge-computing, Osmotic computing, Resource scheduling, Service migrations",
author = "Vishal Sharma and Jayakody, {Dushantha Nalin K.} and Marwa Qaraqe",
year = "2020",
month = "1",
doi = "10.1016/j.future.2019.09.008",
language = "English",
volume = "102",
pages = "723--737",
journal = "Future Generation Computer Systems",
issn = "0167-739X",
publisher = "Elsevier",

}

TY - JOUR

T1 - Osmotic computing-based service migration and resource scheduling in Mobile Augmented Reality Networks (MARN)

AU - Sharma, Vishal

AU - Jayakody, Dushantha Nalin K.

AU - Qaraqe, Marwa

PY - 2020/1

Y1 - 2020/1

N2 - Resources and services between the servers in Mobile Augmented Reality Networks (MARN) are tedious to manage. These networks comprise users possessing Augmented Reality (AR)-Virtual Reality (VR) applications. Low latency, robustness, and tolerance are the key requirements of these networks, which can be attained by using near-user solutions such as edge computing. However, management of services and scheduling them to near-user servers in an integrated environment of edge and public/private infrastructure are complex tasks. These require an optimal solution, which can be obtained by using “Osmotic Computing”, that has been recently proposed as a paradigm for the integration of edge and public/private cloud. This paper uses osmotic computing for effectively migrating and scheduling the services between the servers of the different layers. The paper also presents the details on various components that are used for applying osmotic computing to a network followed by core applications, types, service classification, migration, and scheduling through the rules of osmotic game formulated for its operations. The evaluations are conducted on 100,000 requests and the proposed approach shows significant performance with the probability of the error being 0.1 at 55.72% conservation of the energy and memory resources for the entire network despite the increasing number of users. The proposed approach also satisfies the conditions of the joint optimization functions presented in the system model and demonstrates that the system holds true even with varying users, thus, proving its robustness and tolerance against the number of users.

AB - Resources and services between the servers in Mobile Augmented Reality Networks (MARN) are tedious to manage. These networks comprise users possessing Augmented Reality (AR)-Virtual Reality (VR) applications. Low latency, robustness, and tolerance are the key requirements of these networks, which can be attained by using near-user solutions such as edge computing. However, management of services and scheduling them to near-user servers in an integrated environment of edge and public/private infrastructure are complex tasks. These require an optimal solution, which can be obtained by using “Osmotic Computing”, that has been recently proposed as a paradigm for the integration of edge and public/private cloud. This paper uses osmotic computing for effectively migrating and scheduling the services between the servers of the different layers. The paper also presents the details on various components that are used for applying osmotic computing to a network followed by core applications, types, service classification, migration, and scheduling through the rules of osmotic game formulated for its operations. The evaluations are conducted on 100,000 requests and the proposed approach shows significant performance with the probability of the error being 0.1 at 55.72% conservation of the energy and memory resources for the entire network despite the increasing number of users. The proposed approach also satisfies the conditions of the joint optimization functions presented in the system model and demonstrates that the system holds true even with varying users, thus, proving its robustness and tolerance against the number of users.

KW - Augmented reality

KW - Edge-computing

KW - Osmotic computing

KW - Resource scheduling

KW - Service migrations

UR - http://www.scopus.com/inward/record.url?scp=85072586157&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85072586157&partnerID=8YFLogxK

U2 - 10.1016/j.future.2019.09.008

DO - 10.1016/j.future.2019.09.008

M3 - Article

AN - SCOPUS:85072586157

VL - 102

SP - 723

EP - 737

JO - Future Generation Computer Systems

JF - Future Generation Computer Systems

SN - 0167-739X

ER -