Dissertation > Excellent graduate degree dissertation topics show

Research on Performance Anomaly Problems Diagnosis in Parallel File Systems

Author: DingZuo
Tutor: WangFang
School: Huazhong University of Science and Technology
Course: Computer System Architecture
Keywords: parellel file system performance anomaly problems diagnosed metrics collected peer-comparison
CLC: TP316.4
Type: Master's thesis
Year: 2013
Downloads: 3
Quote: 0
Read: Download Dissertation

Abstract


Parallel file system can experience performance anomaly problems that can be hardto diagnose an isolate. Often, the most interesting and trickiest problems to trace are notnecessarily the outright crash (fail-stop) failures, but rather those that result in a“limping-but-alive” system, i.e., the system continues to operate, but with degradedperformance. Targeting the “limping-but-alive” problem diagnosis in parallel file systemsused for high performance cluster computing (HPC), puting forward a black-box method.By observing the behavior of stripe-based parallel file systems, find that they havesome characteristics in common. Under a performance anomaly in cluster, performancemetrics exhibit observable anomalous behavior on the culprit servers. Base on that, putingforward a diagnosis method by peer-comparison. This approach uses the Kullback-Leibler(Kl) divergence to compare the performance metrics of all the servers, then indicting thefaulty node. By further analysis, it can also find the root-cause.The method diagnoses different performance problems by identifying, gathering andanalyzing OS-level, back-box performance metrics on every node in the cluster. Usingpeer-comparison diagnosis approach compares the statistical attributes of these metricsacross I/O servers, to identify the anomalous node. This method avoids any modificationto the source codes by being transparent to applications. This approach works commonlyacross stripe-based parallel file system. And the approach has good accuracy.At last, the approach is demonstrated by injecting performance problems into the filesystem in both Capfs and Lustre clusters.

Related Dissertations

  1. A Study on the Industry Standardization, the Yardstick Competition Theories and the Progress of the Natural Monopoly Industry,F203
  2. Transplant of Windows CE Operation System Based on ARM9,TP316.7
  3. Kernel Analysis and Application Research of Embedded Real-time Operating System MQX,TP316.2
  4. Linux kernel process scheduling algorithm analysis, research and improvement,TP316.81
  5. Improvement and Implementation μC/OS- Ⅱ real-time operating system kernel analysis and critical technologies,TP316.84
  6. Design and implementation of vehicle transport of dangerous goods monitoring terminal based uC/OS- Ⅱ,TP316.84
  7. Management and Collaboration of Applications in Virtual Desktop System,TP316.7
  8. A Web-based Desktop Virtualization System,TP316.7
  9. Application of VxWorks Operating System and FPGA Technology in Display Control Simulator,TP316.2
  10. Research on Security of Mac Os X Applications,TP316
  11. Embedded real-time operating system ARTs-OS in the TCP / IP protocol stack development,TP316.2
  12. ARM platform to achieve the Linux kernel virtual machine technology research,TP316.81
  13. Based FMS02 Tablet PC prototype of the Linux kernel and driver architecture,TP316.81
  14. Based on Embedded Linux remote desktop technology research and implementation,TP316.81
  15. Virtual desktop support mechanisms of the external device,TP316.7
  16. Android OS Storage Technology,TP316
  17. In the Android system to fight the micro- experimental study,TP316
  18. Study and Implementation of Performance Optimization SAN Cluster File System,TP316.7
  19. Research on Operating System Scheduling Architecture and Algorithm,TP316.81
  20. Research on Aspect-Oriented Modeling and Implementation Method for Real-time System,TP316.2
  21. Information Flow Control Model in Distributed Systems,TP316.4

CLC: > Industrial Technology > Automation technology,computer technology > Computing technology,computer technology > Computer software > Operating system > Distributed operating systems, parallel -type operating system
© 2012 www.DissertationTopic.Net  Mobile