Advanced

Performance modelling and simulation of the Mobile Cloud Network

Tärneberg, William LU (2015)
Abstract
The Mobile Cloud Network is an emerging distributed cloud infrastructure paradigm that attempts to accommodate the evolution of application’s execution paradigms and how content is distributed. The Mobile Cloud Network employs a distributed cloud infrastructure with data centres of varying capacity, embedded in the core and access networks. Resources are thinned over the network and arguably decay in capacity to- wards the capillaries of the network. For an operator of such an infrastructure, cost and infrastructure integrity is their primary concern. Although these aspirations seem triv- ial, achieving them in a highly dynamic and heterogeneous distributed infrastructure, successfully at scale is non trivial. In this work, we have focused... (More)
The Mobile Cloud Network is an emerging distributed cloud infrastructure paradigm that attempts to accommodate the evolution of application’s execution paradigms and how content is distributed. The Mobile Cloud Network employs a distributed cloud infrastructure with data centres of varying capacity, embedded in the core and access networks. Resources are thinned over the network and arguably decay in capacity to- wards the capillaries of the network. For an operator of such an infrastructure, cost and infrastructure integrity is their primary concern. Although these aspirations seem triv- ial, achieving them in a highly dynamic and heterogeneous distributed infrastructure, successfully at scale is non trivial. In this work, we have focused on modelling the dy- namics of the Mobile Cloud Network infrastructure and what methods the operator of a Mobile Cloud Network can employ to minimise the cost of operating the infrastruc- ture. The reduced parameter model, reveals the non-linear nature of the Mobile Cloud Network. The system’s performance is evaluated using a cost function that encom- passes the cost of executing the applications, the cost of the incurred link usage, and the expected performance of the application. Furthermore, simulations reveal that the proposed methods can achieve near optimal placement at reduced time-complexity. (Less)
Please use this url to cite or link to this publication:
author
supervisor
organization
publishing date
type
Thesis
publication status
published
subject
pages
139 pages
ISBN
978-91-7623-580-5
language
English
LU publication?
yes
id
8fc7322b-6142-44e5-af15-1431565c381d (old id 8523921)
date added to LUP
2016-01-19 15:49:11
date last changed
2016-09-19 08:45:18
@misc{8fc7322b-6142-44e5-af15-1431565c381d,
  abstract     = {The Mobile Cloud Network is an emerging distributed cloud infrastructure paradigm that attempts to accommodate the evolution of application’s execution paradigms and how content is distributed. The Mobile Cloud Network employs a distributed cloud infrastructure with data centres of varying capacity, embedded in the core and access networks. Resources are thinned over the network and arguably decay in capacity to- wards the capillaries of the network. For an operator of such an infrastructure, cost and infrastructure integrity is their primary concern. Although these aspirations seem triv- ial, achieving them in a highly dynamic and heterogeneous distributed infrastructure, successfully at scale is non trivial. In this work, we have focused on modelling the dy- namics of the Mobile Cloud Network infrastructure and what methods the operator of a Mobile Cloud Network can employ to minimise the cost of operating the infrastruc- ture. The reduced parameter model, reveals the non-linear nature of the Mobile Cloud Network. The system’s performance is evaluated using a cost function that encom- passes the cost of executing the applications, the cost of the incurred link usage, and the expected performance of the application. Furthermore, simulations reveal that the proposed methods can achieve near optimal placement at reduced time-complexity.},
  author       = {Tärneberg, William},
  isbn         = {978-91-7623-580-5},
  language     = {eng},
  note         = {Licentiate Thesis},
  pages        = {139},
  title        = {Performance modelling and simulation of the Mobile Cloud Network},
  year         = {2015},
}