Study on reliability optimization problem of computer By Dharmendra Singh[Srm unuversity] Delhi Ncr

Submitted By:
Dharmendra Singh
Reg.no-RA1511008030061
IT’A’
Submitted to:
Mrs Kanika garg
Faculty IT department
Seminar-II, 15IT381L

Contents
 Introduction
 Literature survey
 Problem statements
 Objects
 Technique & tools
 Experimental Result
 Conclusion
 References

Introduction
 Reliability problems produced in the 1920s. At that time, people just have
a preliminary understanding that the problem of reliability.
 With the rapid development of computer network, the problem of
reliability of computer network is becoming more and more attention of
builders, users and the network designer.
 Reliability optimization and design of computer network is proposed
firstly by Boecsh, the main issues related to the computer network
reliability optimization and improvement of computer network topology.
 Study on the reliability of computer network is increasingly deepening,
and constantly open up new field potential. This phenomenon strongly
suggest that, on the computer network reliability will gain greater impetus,
its scope of application will be more expanded, in practical applications
will achieve more fruitful results

Literature survey
 Transmission expansion by branch-and-bound integer
programming with optimal cost-capacity curves, Power paratus and
Systems.
 Two factors distinguish between a traditional design and a re-
design concept for a supply chain network design.
 The first aspect is the state of the properties potential sites (i.e., the
number, capacity level of the facilities, transportation and
warehousing costs) that affect the optimal configuration of the
supply chain network.
 Furthermore, every change in the network state (disruption,
changing in costs, etc.) with respect to improve the service level of
customers need substantial capital investment and has a long-
lasting effect on the efficiency of the supply chain activities .

 Second, a re-design concept should be changed gradually the state
of existing facilities (phase out, capacity transfer, etc.). For this
reason, a re-design is more complex compared with a traditional
design of a supply chain network. The re-design aspect in the
supply chain network has not received as much attention as
traditional aspect in the literature.
 Due to the numerous researches implemented in the traditional
design, we only present the related recent studies in re-design,
reliability and uncertainty issues in supply chain network.

Task allocation for maximizing
reliability of distributed computer
systems
 For distributed systems, system reliability is defined as the probability that
the system can run an entire task successfully.
 When the system's hardware configuration is fixed, the system reliability
is mainly dependent on the software design.
 The task allocation problem is addressed with the goal of maximizing the
system reliability. A quantitative problem model, algorithms for optimal
and suboptimal solutions, and simulation results are provided and
discussed.

Reliable path for virtual endoscopy:
ensuring complete examination of
human organs
 Virtual endoscopy is a computerized, noninvasive procedure for detecting
anomalies inside human organs.
 Several preliminary studies have demonstrated the benefits and
effectiveness of this modality. Unfortunately, previous work cannot
guarantee that an existing anomaly will be detected, especially for
complex organs with multiple branches.
 In this paper, we introduce the concept of reliable navigation, which
ensures the interior organ surface is fully examined by the physician
performing the virtual endoscopy procedure.
 To achieve this, we propose computing a reliable fly-through path that
ensures no blind areas during the navigation

Optimum design of IC
power/ground nets subject to
reliability constraints
 The authors formulate and solve the problem of sizing power/ground (p/g)
nets in integrated circuits composed of modules, where the nets are routed
as trees in the channels between the modules.
 Constraints are developed to maintain proper logic levels and switching
speed, to prevent electro migration, and to satisfy certain design rule
requirements.
 The objective is to minimize the area of the p/g nets subject to these
constraints. An optimization technique tailored to this problem is
developed.
 The technique solves the problem more efficiently than the steepest
descent method and Newton's method. Several case studies are presented.

Algorithm
 Genetic algorithm (Gneetic Algorhitm, referred to as GA) is an
effective method to solve the optimization problem. The first it is
inspired by the J.Holand in 1975 by evolutionism, and puts forward
.
 GA is based on the "survival of the fittest" is a highly parallel,
random and adaptive optimization algorithm, it shows the solution
of the problem as "survival of the fittest chromosomes during",
through the "chromosome" group of generations of evolution,
including reproduction, crossover and mutation, eventually
converge to the "best adapted to their environment" the individual,
thus obtains the optimal solution of the problem or satisfactory
solution. At present, with the development of computer technology.
 GA has been paid more and more attention, and in machine
learning, pattern recognition, image processing,

 F(x)=(x-1)/(pop-size-1) Where: x-the position of individual in the cost
sequencing.
 (1<=x<= Pop- size) The evolutionary computation (option).
 This paper by the way of “runner selection”, the basic idea is that each
gene selection probability proportional to its fitness.
 For fitness for human gene, the choice probability ruler is calculated
according to the following formula

Problem statements
 The problem of reliability of computer network is becoming more and
more attention of builders, users and the network designer .
 The reliability of computer network has become a key technical indicators
to measure the comprehensive performance of computer network .
 The main issues related to the computer network reliability optimization
and improvement of computer network topology.
 Early research on computer network reliability, it is mainly based on
computer network connection and cohesion as the measure of the
reliability of computer network .

Objects
 We propose a bi-objective two-stage stochastic mixed-integer
linear programming.
 The importance of considering reliability concepts is considered
and modeled.
 Two type of delivery lead times has taken into account under
reliability.
 An industrial case study is used to show the applicability of the
proposed optimization model.
 In order to obtain Pareto optimal solutions, a novel version of
constraint method is applied.

Technique & tools
 Bathtub curve
 Configuration model
 Critical item list
 Decision trees
 Contract for reliability

Experimental results
 This paper cite optimization simulation example calculation of a
computer network reliability.
 Computer network link cost model to see the model, points N=6
node computer network, computer network node reliability
constraint constant,, the number of iterations of genetic operation is
100 times.
 The cost matrix of computer network link and computer network
reliability matrix ,respectively as follows.

The Simulation Graphs of the Cost
of Computer Network Link

The Simulation Graphs of the
Reliability of Computer Network

Conclusion
 The problem of reliability optimization design of computer network is a complex
optimization problem .
 Genetic algorithm, neural network, fuzzy neural network intelligent algorithm
provides a new idea and approach for solving this kind of problem
 The rapid development of the computer network, increasing factors affecting the
reliability of computer network promote hand, on the other hand, put forward
higher requirements "therefore, in order to measure the impact of computer network
reliability index system of computer network reliability is mentioned in the paper
and the factors of computer network reliability index system should be perfected,
enriched and improved .

References
 [1] S. T. Y. Lee, K. L. Hicks and E. Hnyilicza, “Transmission expansion by
branch-and-bound integer programming with optimal cost-capacity curves,
Power Apparatus and Systems, IEEE Transactions, vol. 5, (1974), pp.
1390-1400.
 [2] M. G. Rabbat, R. D. Nowak and J. A. Bucklew, “Generalized
consensus computation in networked systems with erasure links”, Signal
Processing Advances in Wireless Communications, IEEE 6th Workshop,
IEEE, (2005), pp. 1088-1092.
 [3] S. Kartik and C. S. R. Murthy, “Task allocation algorithms for
maximizing reliability of distributed computing systems”, Computers,
IEEE Transactions, vol. 46, no. 6, (1997), pp. 719-724.
 [4] B. Dengiz, F. Altiparmak, A. E. Smith, “Efficient optimization of all-
terminal reliable networks, using an evolutionary approach”, Reliability,
IEEE Transactions, vol. 46, no. 1, (1997), pp. 18-26.
 [5] T. He, L. Hong and D. Chen D, “Reliable path for virtual endoscopy:
ensuring complete examination of human organs”, Visualization and
Computer Graphics, IEEE Transactions, vol. 7, no. 4, (2001), pp. 333-342.

Study on reliability optimization problem of computer By Dharmendra Singh[Srm unuversity] Delhi Ncr

More Related Content

What's hot (20)

Similar to Study on reliability optimization problem of computer By Dharmendra Singh[Srm unuversity] Delhi Ncr (20)

Recently uploaded (20)

Study on reliability optimization problem of computer By Dharmendra Singh[Srm unuversity] Delhi Ncr