Database Reference
In-Depth Information
with the challenge of delivering optimised and streaming content to application running
on devices including tablets and smart-phones while ensuring high-speed access and
superior performance. The major challenges that the emerging applications bring to
the future internet [
23
] include the requirements of: (1) higher scalability, (2) higher
capability, (3) higher quality of service (QoS), (4) stronger interactivity, (5) dealing with
heterogeneity (e.g., device, network and application) and (6) security. Content delivery
networks (CDNs) are often required to face the data deluge to ef
ciently and securely
distribute content to a large number of online users. The growth of related technologies
such as accelerated web performance, rich media content streaming, IPTV, management
and delivery of user generated content over the last decade has led to the signi
cant
adoption of CDNs. Cisco has estimated that over half of the internet traf
c generated
will be carried out by content delivery networks by 2018.
A CDN is a distributed network of servers and
le storage devices that replicates
content/services (e.g.
les, video, audio etc.) on a large number of surrogate systems
placed at various locations, distributed across the globe. CDNs are highly flexible and
aims to improve the quality and scalability of the services offered over the Internet by
reducing the latency and ef
ciency of delivering contents to clients. The CDN maxi-
mises the bandwidth for accessing to data from clients throughout the network by
strategically placing content replica(s) at geographically distributed locations. The
concept of a CDN was conceived during the early days of Internet. By the end of
1990
'
s before CDNs from Akamai and other commercial providers managed to deliver
Web content (i.e., web pages, text, graphics, URLs and scripts) anywhere in the world,
and at the same time meet the high availability and quality expected by their end users.
Today, Akamai [
22
] delivers between
fteen to thirty percent of all Web traf
c,
reaching more than 4 Tb per second.
In today
s dynamic Internet landscape, it is more important than ever for content
and service providers to understand the requirements and demands of users. For
instance, consider a video distribution services such as Netflix, YouTube and Quickflix.
When delivering video content to geographically distributed subscribers, the video
experience can vary depending on the delivery path to the subscriber. Studies [
20
,
21
]
show that, the sensitivity of subscribers to video quality issues can greatly impact the
subscriptions to the services offered by the video distribution service providers.
Cloud computing is an emerging computing model where a myriad of virtualized
ICT resources are exposed as web utilities, which can be invoked and released in an on-
demand fashion [
24
,
25
]. The concept of cloud computing is an immediate extension of
many well researched domains such as virtualisation, distributed, utility, cluster, and
grid computing. The most comprehensive, widely used and referred de
'
nition of cloud
computing in the literature is presented in [
16
]. It de
nes cloud computing as
“
A model
for enabling convenient, on-demand network access to a shared pool of con
gurable
computing resources (e.g., networks, servers, storage, applications, and services) that
can be rapidly provisioned and released with minimal management effort or service
provider interaction
”
. A number of public cloud providers including Amazon Web
Services (AWS), Microsoft Azure, Salesforce.com and Google App Engine have been
emerged to be very successful in the recent past. The advent of virtualization has led to
the transformation of traditional data centres into flexible cloud infrastructure.
Search WWH ::
Custom Search