Dissertation > Excellent graduate degree dissertation topics show

Grid-based neural computing platform design and implementation of resource allocation

Author: LaiZhuo
Tutor: YangJianGang
School: Zhejiang University
Course: Computer Applications
Keywords: Neural Network Grid Neural computing platform Resource allocation
CLC: TP393.09
Type: Master's thesis
Year: 2004
Downloads: 148
Quote: 0
Read: Download Dissertation

Abstract


Neural network research is quite extensive, with a variety of scientific developments in the field is closely related to the current speech recognition, character recognition, imaging techniques, pattern recognition and classification, signal process, process control and optimization and other fields for many applications. It is because more and more applications require the support of neural networks, we set out to develop a neural computing platform to achieve a variety of neural network models, such as BP model, Hopfield model, Boltzmann machine model, ART model, BAM model, genetic algorithms and so on. And according to the needs of users to combine multiple models will be calculated. But because the data processed neural computation relatively large, so in order to improve processing speed, we introduced a grid (Grid Computing) the structure of thinking, a distributed system architecture. This paper presents a platform resource allocation system design and implementation. Systems Globus project information services - Element MDS directory services to discover resources. It is primarily a grid computing environments based information services framework for grid computing environment, a huge number of resources, geographically distributed dynamic characteristics. MDS content includes resource (service) discovery, resource (service) description and resource (services) monitoring and updating. Through MDS, the system can obtain real-time resource information to meet their needs. And a resource allocation method and the conventional type, the present system uses a data model called Classified Advertisement. This model is very flexible and easy to extend. It is a characteristic of the resource queries to the data model encapsulated in order to express the property. Constrain property which is the machine to the task, or the task of the machine limits. The task is defined quality of service to be reflected in the Rank property. In this way, different users according to their actual situation Qos standards define different, more reasonable. Classad Another feature is that it also allows resource providers to define their own use policies do not meet the resource usage limits itself to those tasks, resources can be rejected. This is for all of the resources distributed environment, given the right to choose the machine, so that the whole system is more flexible. Article is divided into five parts, the first part describes the development of neural networks and grid technology is proposed neural computing platform designed significance; second part describes the resource allocation system design and implementation, including the main module, the mapping algorithm and how to activate a specific task ; The third part introduces the Globus project MDS use; detailed description Classified Advertisement fourth part of the data structure. The final part of the fifth and sixth running instance with the article summary.

Related Dissertations

  1. Development of the Platform for Compressor Optimization Design and Aerodynamic Optimization Design in the Transonic Compressor,TH45
  2. Grid-Side Converter Control and Wind Turbine Emulator in Direct Drive Wind Power System,TM46
  3. Research on Feature Extraction and Classification of Tongue Shape and Tooth-Marked Tongue in TCM Tongue Diagnosis,TP391.41
  4. Study on Virtual Detector of Infrared Hyper-Spectral Image,TP391.41
  5. Research on Visual Servo System of Mechanical ARM,TP242.6
  6. Design and Realize of Family Cleaning Robot Path-Coverage System,TP242
  7. Design of Weapon Detection Device Control System,TP183
  8. Municipal tourism land use planning environmental impact assessment,X820.3
  9. Study on Taste Characteristic of Taste Peptide Enzymatic Production from Oyster Base on A Neural Network Method,TS254.4
  10. The Grid-Connected Wind-solar Hybrid Generation System and Maximum Power Point Tracking,TM61
  11. Research of Scheduling Algorithm Based on Hybrid Adaptive Genetic Algorithm in Computing Grid,TP393.09
  12. BP network optimization based on genetic algorithm optimization of the biodiesel process,TE667
  13. Fire Fighting System Research for the Offshore Platform,U698.4
  14. The Establishment of Grid Platform on Agricultural Supply Chain System,S126
  15. Research on Inspection Technology of Dehydrated Garlic Slice Based on Computer Vision,TP391.41
  16. Research and Application of Safety Assessment Method Based on Neural Network,X937
  17. The Design of the Coal Mine’s Safety Evaluation System Based on Wavelet Neural Network,TD79
  18. Research of VRLA Battery On-line Montoring and Contorl System,TP277
  19. The Identification of Fault Type in Transmission Lines Based on Neural Network,TP183
  20. Enterprise Security Benefit Evaluation and Development Strategies,F272;F224.5
  21. Mine Risk Information Integration and Intelligent Early Warning,X936

CLC: > Industrial Technology > Automation technology,computer technology > Computing technology,computer technology > Computer applications > Computer network > General issues > The application of computer network
© 2012 www.DissertationTopic.Net  Mobile