When pressing "D" within MTR, one sees a "Last n pings" side-scrolling output. At the bottom is a scale with the different max-times. I understand how this relates to the scrolling data.

Scale: .:1 ms 1:4 ms 2:10 ms 3:17 ms a:28 ms b:40 ms c:54 ms >

How is this scale calculated? Periodically, it will change to reflect, I assume, new data. My guess is that it has to do with distributions of response times and that each level would represent a certain percentage.

I'd like to be able to look at the scale and say "95% of packets are returned within 28ms" for example.

How is the scale arrived at?

migrated from serverfault.com Nov 18 '18 at 22:19

This question came from our site for system and network administrators.

  • 1
    Look at the source. – Ipor Sircer Nov 14 '18 at 8:23
  • 1
    Thank you for that, @IporSircer. I can understand RTFM, but RTFSC is a bit much. – Philip Nov 14 '18 at 8:32

As far as I understand the source it is a square law.

Experimentally, with 7 scale items

  • take the longest ping time
  • take the square root and divide by 7² (for instance 166ms/49=3.388)
  • then items for N=1..6 are N²*3.388

Take in account roundoff errors, for instance the displayed 166ms could be 165.5...

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.