Font Size: a A A

Research Of A Visual Monitor For Dynamic Behaviors Of The Parallel Programme And Research Of Parallel I/O

Posted on:2004-06-22Degree:MasterType:Thesis
Country:ChinaCandidate:X W LiFull Text:PDF
GTID:2168360092493105Subject:Applied Mathematics
Abstract/Summary:PDF Full Text Request
With the fast development of high-performance parallel computers, the application of parallel computing is becoming more extensive. The Cluster System based on distributed-storage are been paying attention to day by day. Users using the Cluster System need eagerly for a tool for performance analysis and monitor , which can show the characters of parallel programmes with a visual way.In this paper , we introduced briefly the technology of Cluster System, the proposal and development of the tools for performance analysis and monitor, the classification and current developing situation of the common performance analysis tools. We also gave a profile of the research and application of parallel I/O, including Parallel I/O file systems, the classification and compare of the common Parallel I/O systems, and a actual example of using parallel I/O to collect data for the tool of performance analysis and monitor.The entire design process and all of the key technologies of the PPDBM (Parallel Programme Dynamic Behavior Monitor) are illustrated in the paper, a performance analysis and monitor tool based on the Cluster System. It includes the implement of the software events recorder , the dynamic visualization of the status of the processes, the status of the CPUs and Memories of the running nodes, and the graphical show of the analyzed data of parallel accelerate rate, parallel efficiency and computing-communication ratio.At last, a classical parallel programme of computing the approximate value of n was run with different number of nodes and different scale of question on the self-established Linux cluster system, and the process status and the performance results were monitored by the PPDBM at the same time.
Keywords/Search Tags:Cluster System, Parallel I/O, performance analysis and monitor tools, parallel computing environment, software events recorder
PDF Full Text Request
Related items