A SCALABLE OVERLAY
MULTICAST ARCHITECTURE FOR LARGE SCALE APPLICATION
abstract
In this paper,
we propose a Two-tier Overlay Multicast Architecture (TOMA) to provide scalable
and efficient multicast support for various group communication applications.
In TOMA, Multicast Service Overlay Network
(MSON) is advocated as the backbone service domain, while end users in access
domains form a number of small clusters, in which an application-layer
multicast protocol is used for the communication between the clustered end
users.
TOMA is able
to provide efficient resource utilization with less control overhead,
especially for large-scale applications. It also alleviates the state
scalability problem and simplifies multicast tree construction and maintenance
when there are large numbers of groups in the network.
To help MSON
providers efficiently plan backbone service overlay, we suggest several
provisioning algorithms to locate proxies, select overlay links, and allocate
link bandwidth. Extensive simulation studies demonstrate the promising
performance of TOMA.
Analysis on Existing Networks:
The
existing network level multicasting is used to send the packets from source to
all the destination nodes. Each intermediate node which maintain the table and
search the destination address in the
table if the destinations address available then the relay node send the packets to destination or else it
send the packets to nearest neighboring node. In network level multicasting,
·
Network in charge of selecting relay
node.
·
Relay node selection affects user
performance.
·
It
yields sub-optimal user performance.
·
Reduces the packet delivery ratio
Deficiency
in Existing approach:
– Low packet delivery
ratio.
– High forwarding ratio
– High delay.
– Multicast redundancy.
Idea
on proposed network:
In the proposed multicast system, the packets are
sending very faster without any dealy.There is one type of algorithm is used
here to improve all the drawbacks of exiting approach. They are
·
Double Covered multicast
algorithm
Double Covered multicast algorithm:
The proposed double-covered multicast (DCB) algorithm works
as follows: When a sender multicasts a packet, it selects a subset of 1-hop
neighbors as its forwarding nodes to forward the packet based on a greedy
approach. The selected forwarding nodes satisfy two requirements:
1) They cover all the sender’s 2-hop
neighbors, and
2) The sender’s 1-hop neighbors are either
forwarding nodes or no forwarding nodes covered by at least two forwarding
nodes (e.g., once by the sender itself and once by one of the selected
forwarding nodes).
After
receiving a new multicast packet, each forwarding node records the packet,
computes it’s forwarding nodes, and multicasts the packet as a new sender. The
retransmissions of the forwarding nodes are overheard by the sender as the
acknowledgement of the reception of the packet.
The
no forwarding 1-hop neighbors of the sender do not acknowledge the receipt of
the multicast. The sender waits for a predefined duration to overhear the multicast
from its forwarding nodes. If the sender fails to detect all its forwarding
nodes retransmitting the packet it will resent the packet.
We propose a simple multicast
algorithm, called double covered multicast (DCB), which takes advantage of
multicast redundancy to improve the delivery ratio in the environment that has
rather high transmission error rate. Only a set of selected nodes will forward
the multicast message.
The selected nodes, called
forwarding nodes, meet the following two requirements: 1) they cover the
sender’s 2-hop neighbor set, and 2) they cover the sender’s 1-hop no forwarding
neighbors at least twice. Also, the retransmissions of the forwarding nodes are
received by the sender as the acknowledgement of their reception of the
packet.1 No forwarding neighbors do not acknowledge the reception of the
multicast.
If the sender fails to detect all its
forwarding nodes’ retransmissions, it repeatedly resends the packet until it
detects that all the retransmissions or the maximum number of retries is
reached.
The proposed algorithm has
many merits, such as balancing the average retransmission redundancy, avoiding
both the multicast storm problem and the ACK implosion problem, recovering the
transmission error locally, and increasing the multicast delivery ratio in a
high transmission error rate environment.
Simulation results show that the algorithm
provides high delivery ratio, low forwarding ratio, low overhead, and low
end-to-end delay for a multicast operation under a high transmission error rate
environment.
Limitations
•
Network
multicasting is suboptimal for user
performance
•
In
this IP unicasting and application layer multicasting is suitable only for a small
network.
•
In
large network, it is difficult to identified and maintained the root path
information by source
PROPOSED SYSTEM:
•
In
the proposed overlay multicasting, the end users pick their own routing path to
send the packets from source to destination.
•
There
are two types of multicasting, they are
i)
Application layer multicasting
ii)
Overlay multicast
o
Optimize
their own performance goals – Not considering system-wide criteria.
•
Studies
based on small scale deployment show it improves performance.
q Application layer
multicast
§ no need to use any switches and
interdomain routers
§ Optimize their own performance goals – Not considering system-wide
criteria.
Application
Layer Multicasting:
o
Studies based on small scale
deployment show it improves performance.
o
In
applications layer multicasting destination node can directly choose the source
path.
o
It
results in overlay multicasting; the data delivery ratio is increased.
o
Reduces
the system resources
o
Simplify
the maintenance
o
Drawback
o
It
is not suitable for large scale applications
System Requirements
Software Requirements
Language : Java1.5
Front End Tool :
Swing,
Operating System : Windows XP.
Back End : Sql Server
Hardware Requirements
PROCESSOR : Intel Pentium III Processor
Random Memory : 128MB
Hard DISK : 20GB
Processor SPEED:
300 min
No comments:
Post a Comment