Evaluation Of Load Balancing Algorithms And Internet Traffic Modeling For Performance Analysis
This thesis presents research and study of load balancing algorithms and the analy-sis of the performance of each algorithm in varying conditions. The research also covers a study of the characteristics of Internet traffic and its statistical properties. The network workload models that were implemented in the simulation program were derived from the many works already published within the Internet community. These workload models were successfully implemented and statistical proof is given that they exhibit characteris-tics similar to the workloads found on the Internet. Finally, this thesis compares and con-trasts the differences between stateless server selection methods and state-base selection methods with the different algorithms studied.
The research and results of the topic of Load Balancing and Internet traffic model-ing are presented in this thesis.
Load Balancing is a form of system performance evaluation, analysis and optimi-zation, which attempts to distribute a number of logical processes across a network of pro-cessing elements. There have been many algorithms and techniques that have been developed and studied for improving system performance. In early research in the field of computer science, the main focus for improving performance was to develop algorithms and techniques to optimize the use of systems with limited and expensive resources for scientific computing and information systems. Later, there was an emphasis on how to network groups of computers or workstations and then share the resources among work-groups. More recently there has been a tremendous increase in the popularity of the Inter-net as a system for sharing and gathering information. The use of the Internet has been increasing at a tremendous rate and there always has been a concern among those in the Internet community that enough resources will be available to provide the expected qual-ity of service that is received by its users.
The process of “balancing”, “sharing”, “scheduling” or “distributing” work using a network of computers or a system of multiple processing elements is a widely studied subject. This paper will focus on recent works that have been written regarding today’s com-puting environments.
In this thesis, we will look at the art of load balancing and how it can be applied to distributed networks and more specifically the Internet. Initially, the focus of the research for this thesis was on development and analysis of algorithms that minimized the amount of messaging or probing that is required for determining the current workload of a set of processing elements, such as a group of replicated web servers. These algorithms were to compare stochastic based methods of estimating server workloads, with more intrusive methods of messaging and probing. During the process of developing the simulator to be used for evaluating the algorithms in this study, tasks related to modeling network work-loads, and the workloads related to the Internet in particular, were identified to be crucial to the research in this area and as a result of this effort, network modeling has become a significant portion of this thesis.
The thesis is divided into the following chapters. Chapter 2 will discuss the con-cepts and research related to the subject of Load Balancing. Chapter 3 will discuss the issues related to modeling network traffic and in particular Internet traffic. Chapter 4 describes the Network Simulator used to evaluate the load balancing algorithms. Chapter 5 is the Experimental Design of the network simulations, and Chapter 6 will discuss the results of the experimental simulations. Finally, Chapter 7 will discuss the conclusions based upon the experiments performed in this study.
Finally, in addition to the goals already mentioned, it is the hope of the author that this research can be used as a reference for the continued study in the areas of network performance evaluation, modeling and simulation.
In this thesis, the subjects of Load Balancing and Internet Workload Modeling were presented for the purpose of doing performance analysis of different load balancing algorithms in different environments. Chapter 2 contained the important concepts related to load balancing and the discussion of recent research in that area. In Chapter 3, the details of modeling the characteristics of Internet traffic were discussed. These character-istics were shown to have high statistical variability caused by request inter-arrival rates, file size distributions, server workloads and network workloads. It was also shown that the high variability of these characteristics are statistically self-similar. Self-similarity is a fractal-like behavior where the high variations in workloads exist over a wide range of time scales. Chapter 4 contained the implementation details of the workload models devel-oped from researching this thesis. At the end of Chapter 4, the verification and validation of these models is shown. Chapter 5 described the experimental design and Chapter 6 pro-vided an analysis of the results.
The final conclusion in this thesis is that client response times can be reduced by using an agent, referred to here as a load balance manager that can obtain some knowledge of the system state for selecting processing elements to service requests. In comparison of stateless versus state-based load balance managers, the experimental results showed up to 27 times improvement in response time can be achieved by state-based algorithms avoiding servers that are system bottlenecks over stateless algorithms. In all experiments, both stateless algorithms, round robin and random performed the same. Therefore a random algorithm would be the best choice to use because it should be the easiest to implement. Comparing the state-based algorithms the greedy algorithm showed the best performance in all cases, with the subset algorithm performing better than the stochastic algorithms at low server utilization, and the stochastic algorithms out performing the subset algorithm at higher server utilization. The conclusion that can be drawn here is use the greedy algorithms if it is possible. The results of the experiments in this study are inconclusive for selecting either of the subset algorithm or the stochastic algorithm over the other.
This thesis also showed how to model Internet traffic workloads that can be used for additional research to develop algorithms and protocols to be used in the Internet and distributed networks.
How To Get The Complete Material For “Evaluation Of Load Balancing Algorithms And Internet Traffic Modeling For Performance Analysis“
The Complete Material Will Be Sent to You in Just 2 Steps
Quick & Simple…
Make Payment (Through Transfer) of ₦3,000 to Any of the Account Below
|Acc No: 0811003731|
|Acc No: 1225513212|
|Acc No: 8143831497|
Or CLICK HERE To Pay With Debit Card
|FOR CLIENTS OUTSIDE NIGERIA|
|CLICK HERE To Purchase Material ($15)|
|FOR GHANIAN CLIENTS|
|Make Payment of 80 GHS to 0553978005 | Douglas Osabutey | MTN MoMo|
Send the Following Details on WhatsApp ( 08143831497) After Payment
- Payment Details
- Email Address
- Evaluation Of Load Balancing Algorithms And Internet Traffic Modeling For Performance Analysis
The Complete Material Will Be Sent To Your Email Address After Receiving Your Details
T & C Apply
Need a Different Topic? Perform a Quick Search