Dissertation > Excellent graduate degree dissertation topics show

The Research and Implementation of Fault-tolerant Metadata Cluster Management for PB-scale Storage System

Author: WangYong
Tutor: LiuZhong
School: National University of Defense Science and Technology
Course: Computer Science and Technology
Keywords: Metadata Management High Availability A copy of the directory object Log - based metadata Checkpoint Object Storage Fault-tolerant
CLC: TP333
Type: Master's thesis
Year: 2007
Downloads: 101
Quote: 0
Read: Download Dissertation

Abstract


With the development of computer technology, information technology and the Internet , the high-performance computing , business computing , large-scale data processing , information processing technology has been widely used . The cluster system has been more and more popular because of its high cost performance and better scalability . At the same time , these applications of distributed data storage with a larger capacity , higher performance , higher availability requirements . The emerging object storage structures to take advantage of the existing processing technology , network technology and storage components to provide unprecedented scalability and aggregation throughput , provide the foundation for building a new generation of massively parallel storage system . In this paper, a comprehensive understanding of object storage architecture with existing object storage system based on in-depth study of the fault-tolerant cluster management issues involved in large-scale cluster storage system based on the object storage architecture metadata proposed a novel ideas and solutions. The main contributions are as follows: ( 1) highly available metadata management model based on a copy of the directory object oriented PB -level storage system . Efficient concurrent copy of the directory object placement, update , migration strategy and management mechanism , it is necessary to ensure the reliability and availability of the directory object data , but also increase the aggregate network bandwidth when reading data , to improve the performance of read operations . Markov incentive model quantitative usability analysis . Actual functional testing and performance testing show that the model can effectively guarantee the high availability of storage systems , improve overall access performance of the metadata server cluster . (2) a checkpoint log-based metadata management - oriented way of thinking . Through the implementation of the checkpoint log - based metadata management , both to ensure the effective use of the system storage space , while achieving the fast recovery of the system to ensure the high availability of the system . The practical test show that this method can fully utilize the disk space , and improve the recovery time of the system , to ensure the high availability of metadata to improve the efficiency of the access to the metadata .

Related Dissertations

  1. The Study and Design of High Availability Monitoring Subsystem for Fault Tolerant Computing Systems,TP311.52
  2. The Establishment of Grid Platform on Agricultural Supply Chain System,S126
  3. Research on Security Model for Mobile Agent-Based E-Commerce Environment,TP393.08
  4. Design and Implementation of Authentication Mechanism in Object-based Storage System,TP333
  5. Eucalyptus-based Object Storage System Design and Implementation,TP333
  6. Active Object Storage System Realization Mechanism,TP333
  7. Study on Fault-tolerant Guaranteed Cost Control for Nonlinear Systems,TP13
  8. Research on Fault-Tolerant Parallel Skyline Query Technology in Cloud Computing Environment,TP311.13
  9. Design and Implementation of File System in Object Storage Device,TP333
  10. Study on Robust H_∞ Fault-tolerant Control for Parameter Uncertain Time Delay Systems,TP13
  11. Study on Fault-tolerant Control and Filtering Research for Uncertain Systems,TP13
  12. Fault-Tolerant Control for Descriptor System with Time-delay,TP13
  13. Researchs on Fault-tolerant Routing Algorithms in Network on Chip (NOC),TN47
  14. Research on the Reliable Fault-Tolerant Routing of Hypercube Multicast,TN915.02
  15. The Study of Fault Tolerant Control Method,TP13
  16. Object-based storage system design and implementation,TP333
  17. The Research of Fault-Tolerant Techniques for Parallel/Distributed Network Simulator PDNS,TP302.8
  18. Based on Measurement of Satellite Formation Flying Collision Avoidance Study,V556
  19. Research and design of fault-tolerant flight control computer,TP302.8
  20. PIPE- based usability evaluation tools for research,TP302.8
  21. SSH-based composition framework for research and application of online marking system,TP311.52

CLC: > Industrial Technology > Automation technology,computer technology > Computing technology,computer technology > Electronic digital computer (not a continuous role in computer ) > Memory
© 2012 www.DissertationTopic.Net  Mobile