Delay-Based
Network Utility Maximization
ABSTRACT:
It is well known that
max-weight policies based on a queue backlog index can be used to stabilize
stochastic networks, and that similar stability results hold if a delay index
is used. Using Lyapunov optimization, we extend this analysis to design a
utility maximizing algorithm that uses explicit delay information from the
head-of-line packet at each user. The resulting policy is shown to ensure
deterministic worst-case delay guarantees and to yield a throughput utility
that differs from the optimally fair value by an amount that is inversely
proportional to the delay guarantee. Our results hold for a general class of
1-hop networks, including packet switches
EXISTING SYSTEM:
The stability works all use backlog-based
transmission rules, which treat joint stability and utility optimization.
However, work introduces an interesting delay-based Lyapunov function
for proving stability, where the delay of the head-of-line packet is used as a
weight in the max-weight decision. This approach intuitively provides tighter
control of the actual queueing delays.
DISADVANTAGES
OF EXISTING SYSTEM:
1. In
an interesting delay-based Lyapunov function for proving stability,
where the delay of the head-of-line packet is used as a weight in the
max-weight decision. This approach intuitively provides tighter control of the
actual queuing delays.
2. A
single head-of-line packet is scheduled based on the delay it has experienced,
rather than on the amount of additional packets that arrived after it.
3. Use
delay-based rules only in the context of queue stability. To our knowledge,
there are no prior works that use delay-based scheduling to address the
important issue of joint stability and utility optimization.
PROPOSED SYSTEM:
THIS paper considers the
problem of scheduling for maximum throughput utility in a network with random
packet arrivals and time-varying channel reliability. We focus on 1-hop networks
where each packet requires transmission over only one link. At every slot, the
network controller assesses the condition of its channels and selects a set of
links for transmission. The success of each transmission depends on the
collection of links selected and their corresponding reliabilities. The goal is
to maximize a concave and nondecreasing function of the time-average throughput
on each link.
In this paper we use a
delay-based Lyapunov function and extend the analysis to treat joint stability
and performance optimization via the Lyapunov optimization technique from our
prior work. The extension is not obvious. Indeed, the flow control decisions in
the prior work are made immediately when a new packet arrives, which directly
affects the drift of backlog-based Lyapunov functions. However, such decisions do not directly affect the delay
value of the head-of-line packets, and hence do not directly affect the drift
of delay-based Lyapunov functions. We overcome this challenge with a novel flow
control policy that queues all arriving
data, but makes packet dropping decisions just before advancing a new packet to
the head-of-line. This policy is structurally different from the utility
optimization works. This new structure leads to deterministic guarantees on the
worst-case delay of any non dropped packet and provides throughput utility that
can be pushed arbitrarily close to optimal.
ADVANTAGES
OF PROPOSED SYSTEM:
ü It
is important to analyze these delay-based policies because they improve our
understanding of network delay, and because the deterministic guarantees they
offer are useful for many practical systems.
ü while
our deterministic delay guarantees hold for general arrival sample paths, our
utility analysis assumes all arrival processes are independent of each other
(possibly with different rates for each process) and independent and
identically distributed (i.i.d.) over time-slots.
ü The
deterministic delay guarantees we obtain in this present paper are quite strong
and show the advantages of our new flow control structure.
SYSTEM CONFIGURATION:-
HARDWARE CONFIGURATION:-
ü Processor - Pentium –IV
ü Speed - 1.1
Ghz
ü RAM - 256
MB(min)
ü Hard Disk -
20 GB
ü Key Board -
Standard Windows Keyboard
ü Mouse - Two
or Three Button Mouse
ü Monitor - SVGA
SOFTWARE CONFIGURATION:-
ü Operating System : Windows XP
ü Programming Language :
JAVA
ü Java Version :
JDK 1.6 & above.
REFERENCE:
Michael J. Neely, Senior Member, IEEE-
“Delay-Based Network Utility Maximization”- IEEE/ACM TRANSACTIONS ON NETWORKING, VOL. 21, NO. 1, FEBRUARY 2013.
