Execute the jheapprofanalyzer command (class-wise statistical information analysis functionality) to output the total instance size of each class and the number of instances of each class in the time series, assuming multiple extended thread dump files with class-wise statistical information as input files. The files are output in the CSV format.

The following figure shows the flow execute the class-wise statistical information analysis functionality for the CSV output of the information that is acquired as the class-wise statistical information.

Figure 9-16 Flow of CSV output by executing the class-wise statistical information analysis functionality

Using the class-wise statistical information analysis functionality, you can output the information about the instances with large total size and can check the memory usage of only such instances. To output only the instances with large total size, specify the threshold value in -DJP.co.Hitachi.soft.jvm.tools.jheapprofanalyzer.threshold, and then specify in the jheapprofanalyzer command and execute the command. For details about the jheapprofanalyzer command, see jheapprofanalyzer (CSV output of Hitachi class-wise statistical information analysis file) in the uCosminexus Application Server Command Reference Guide.