Tplot is a python and shell mixed package that plots the data gathered from top as a graph/chart for further analysis.
The data can be collected from one machine and plotted on other machine. Basically intended to test and analyse the performance of embedded systems.
A typical setup would be running the shell script to gather information from the embedded Linux machine and plotting it to a decent graph using python program in development machine. But both the machines can be the same.
In embedded Linux machine,
$ ./top-report.sh <Delay for top> <report filename>
<Delay for top> = any integer greater than 1 and less than 10
<report filename> = default is top-report.csv, will be created in
$PWD or provide the filename here.
The above command will generate a csv file containing typical top output, the process related information in comma separated format, which is needed by the plotter python program.
In development Linux machine,
$ python top-plot.py [-h] [-f FILENAME] [-y {VIRT,RES,SHR,MEM,CPU}]
[-x {PID,COMMAND,TIME}] -i ITEM
Plotting of top report generated in work bench.
optional arguments:
-h, --help show this help message and exit
-f FILENAME, --filename FILENAME
The filename from which plot to be generated.
-y {VIRT,RES,SHR,MEM,CPU}, --yfield {VIRT,RES,SHR,MEM,CPU}
The fields to be used for y axis.
-x {PID,COMMAND,TIME}, --xfield {PID,COMMAND,TIME}
The fields to be used for x axis.
-i ITEM, --item ITEM The item to be parsed.
The above command takes the csv file generated from shell script and plots the field it is passed with.