0

I am trying to analyze output of performance benchmark. My spreadseet consists of multiple Linux top outputs spawned every 0,5 s for about a minute. I need to identify keywords that will help me gather all of the information that I need to present it as a graph. The problem is that I don't know how to write a function that would i.e. search for idle word multiple times and each time returned address/reference of a cell containing this string. Any help would be apreciated.

3
  • 1
    It sounds like you have 120 full outputs of top in a running text file. If it were me, I would probably try and use e.g. grep or notepad++ to parse out the lines that are pertinent, massage the results into a tab or comma delimited format and use libreoffice-base to get the data into a table. Graphing would be trivial at that point. I probably would first start by looking to see if there is already a solution to this: firewall packages and other turn-key distributions parse and graph similar data.
    – Yorik
    Commented Oct 16, 2019 at 17:18
  • It is an embedded system and it does have realy good tools which are not apliable to the scenario that I'm testing. I've already done preparsing in Notepad++ so the spreadsheet is well formatted. It can be acutally treated like a set of separated tables, each one containing a seperate top output. Now I would like to find a characteristic string in every table and move a known offset to gather the data automatically. Commented Oct 16, 2019 at 17:25
  • 1
    can you provide a sample parsable line?
    – Yorik
    Commented Oct 16, 2019 at 18:20

0

You must log in to answer this question.

Browse other questions tagged .