An Overview of Cloud Based Content Delivery Networks: Research Dimensions and State-of-the-Art - Transactions on Large-Scale-Data-and Knowledge-Centered Systems XX - page 135

Global Server Load Balancing (GSLB): aims to optimize resource use, maximize
throughput, minimize response time, and avoid overload of any one of the resources.
The capabilities that allow global server load balancing include global awareness and
smart authoritative domain name service (DNS). In GSLB, services nodes are aware of
information and status of other service nodes. This provides intermediate switching
nodes to be globally aware. To make use of the global awareness, intermediate switches
act as smart authoritative DNS, each switching between the best surrogate servers.
DNS-based request-routing: is widely used in the Internet. DNS based request-routing
is also used in many CDNs because of its ubiquity as a directory service. DNS servers
handle the domain name of the desired web site or content. The client initiates a name
lookup in a local DNS server, which is supposed to return the address of a surrogate
server near the client. If local DNS cache misses, it forwards the name lookup to
the DNS root server. DNS root server returns the address of the authoritative DNS
server for the web site. The Authoritative DNS server then returns the address of a
surrogate server near the client based on specialized routing, load monitoring and
Internet mapping mechanism. Finally, the client retrieves the content from the desig-
nated surrogate server. A number of studies have examined and reported the perfor-
mance and effectiveness of DNS [ 33 , 34 ].
HTTP Redirection: takes advantage of the HTTP protocol

s redirection feature. This
mechanism builds on special Web servers that can inspect a client request and chooses
the most suitable surrogate server and redirect the client to those servers. This approach
provides the flexibility of managing replication with

'

ner granularity (e.g., at page

level). However, it does pose signi cant overheads due to the introduction of extra

messages round trips.
URL Rewriting: can be one of the best and quickest ways to improve the usability
and search friendliness. A rewrite engine is software located in a Web application
framework running on a Web server that modi es a web URL ' s appearance. Many

framework users have come to refer to this feature as a cation is

called URL rewriting. For example, request for web sites with images, the router can
rewrite the URLs of the images to point to the best surrogate servers.
Anycasting: is a new routing technology based on the Ipv6. It is a methodology in
which datagrams from a single sender are routed to the topologically nearest node in a
group of potential receivers, though it may be sent to several nodes, all identi

“ Router ” . This modi

ed by the
same destination address. CDNs may use anycast for routing user request to their
distribution centres or DNS.
CDN Peering: is a methodology where clients provide resources; the client can also use
these resources based on their requirements. This means that unlike client-server
systems, the content serving capacity of peer-to-peer networks can actually increase as
more users begin to access the content (especially with protocols such as Bittorrent that
require users to share). This property is one of the major advantages of using P2P
networks because it makes the setup and running costs very small for the original
content distributor. To locate the content in CDN peering [ 34 ], a centralised directory
model, distributed hash table, flooded request model or document routing model can be
used. In centralised P2P le-sharing service, a large server is used to provide directory