IBM®
Skip to main content
    Country/region [select]      Terms of use
 
 
     Home      Products      Services & solutions      Support & downloads      My account     
  IBM Wikis > AIX > Home > Performance Graph Viewer
AIX Log In | Sign Up   View a printable version of the current page.
Performance Graph Viewer
Added by Federico Vagnini, last edited by Federico Vagnini on Nov 25, 2007  (view change)
Labels: 
(None)

Performance Graph Viewer - pGraph.Viewer

pGraph is a Java program designed to read one or more files containing performance data and to produce graphs either interactively or in batch mode. The entire monitoring time frame is handled regardless of its length and sampling time, and user can focus on a specific time period.

No installation is required. Just collect data on the monitored system and copy it where pGraph executable is available. Data can be kept compressed in GZIP format. pGraph automatically detects the data source type and, if input file name end with ".gz", it reads the GZIP format.

pGraph can be easily adapted to read the output of many performance tools. The following formats are currently supported:

The tool is capable of producing graphs related to CPU, memory, Disk, I/O Adapters, Network Adapters and Kernel, depending on the input file content.

Interactive Mode with Java GUI

Batch Mode with HTML browsing 

All graphs contain data related to the same time period and are updated if user changes the time scope. This approach is very useful to get a global view to data and then to zoom to a specific period.

pGraph can read multiple data files at a time and manage them as a whole. There are two available scenarios:

  • Single host, multiple files. When several data files related to the same host are available, they can be stored in the same directory and pGraph will read them all and provide a merged view of data.
  • Multiple hosts, multiple files. The focus is on CPU performance data related to multiple hosts where each file represents one host. pGraph extracts CPU data from each file and creates performance graphs for all hosts, aligned in time. When micro partitions are involved, a sum of global CPU usage is extracted or computed. If all partitions are hosted by the same physical server, shared pool data is provided.

Modes of operation

pGraph has two modes of operation:

  • Interactive Mode to manually select the analisys' time period and the variables to display
  • Batch Mode to automatically create all graphs and an HTML infrastructure for web publishing

Download

pGraph is packaged in a JAR file and requires a Java runtime 1.4 or later. See Modes of operation for details.

This tool is not officially supported by IBM. No warrantee is given or implied, and you cannot obtain help from IBM. It a personal project of the author, Federico Vagnini (vagnini@it.ibm.com). Any comment and suggestion is really appreciated.




Powered by Atlassian Confluence, the Enterprise Wiki. (Version: 2.2.10 Build:#528 Nov 29, 2006)
    About IBM Privacy Contact