News for hackers. Stuff that matters.
Re:Optimizing Linux Memory Management for Low-latency, High-... by introspectif on Tuesday October 08, 2013 @04:10PM (#6517503)
"after rolling out our optimizations, we saw our error rates (ie. the proportion of slow or timed out queries) drop by up to 400%"

There is some good shared knowledge in the post (unlike this comment, to be fair), but what does drop by 400% mean?

If a rate drops by 100% it becomes zero. I get that.

If it increases by 400%, the outcome is slightly ambiguous (do we add 400% for 500% total or do we multiply up to 400% of the original value).

But a rate decreasing by 400% - am I the only person who finds that (not uncommon) expression hard to conceptualize?

Re:Optimizing Linux Memory Management for Low-latency, High-... by erichocean on Tuesday October 08, 2013 @04:25PM (#6517599)
I understood it to mean that it had 1/4 the error rates they were previously seeing.
Re:Optimizing Linux Memory Management for Low-latency, High-... by dllthomas on Tuesday October 08, 2013 @05:16PM (#6517909)
Huh, that's not even one I considered. I guessed 1/5, so the change was 4x the final value. Either way it's wrong - "dropped by X%" should be calculated as (original - final) / original.
Re:Optimizing Linux Memory Management for Low-latency, High-... by apurvamehta on Tuesday October 08, 2013 @04:31PM (#6517630)
This is exactly right :)
Re:Optimizing Linux Memory Management for Low-latency, High-... by caf on Tuesday October 08, 2013 @05:41PM (#6518024)
Then it dropped by 75%.
Re:Optimizing Linux Memory Management for Low-latency, High-... by apurvamehta on Tuesday October 08, 2013 @05:53PM (#6518070)
Yes. It was a blunder. The post has been updated to reflect this.
Re:Optimizing Linux Memory Management for Low-latency, High-... by AsymetricCom on Tuesday October 08, 2013 @04:18PM (#6517551)
Obviously, he gets -300% errors now.
Re:Optimizing Linux Memory Management for Low-latency, High-... by ihsw on Tuesday October 08, 2013 @05:28PM (#6517963)
That's pretty good.