Weekly Edition
Return to the Development page
| |||||||||||||
|
| |||||||||||||
Poettering: Measuring Lock Contention
On his weblog, Lennart Poettering describes a new tool, called "mutrace", for tracking down performance problems caused by mutex contention in applications. "For each mutex a line is printed. The 'Locked' column tells how often the mutex was locked during the entire runtime of about 10s. The 'Changed' column tells us how often the owning thread of the mutex changed. The 'Cont.' column tells us how often the lock was already taken when we tried to take it and we had to wait. The fifth column tell us for how long during the entire runtime the lock was locked, the sixth tells us the average lock time, and the seventh column tells us the longest time the lock was held. Finally, the last column tells us what kind of mutex this is (recursive, normal or otherwise)."
(Log in to post comments)
Poettering: Measuring Lock Contention Posted Sep 15, 2009 19:38 UTC (Tue) by branden (guest, #7029) [Link]
Given my misunderstanding of the headline, I propose:
That the act of measuring lock contention henceforth be defined as "poettering" (likely pronounced "pottering" by those unfamiliar with the rules of German pronunciation); and
That the unit of lock contention (derviation TBD) henceforth be called the "lennart".
Poettering: Measuring Lock Contention Posted Sep 15, 2009 21:26 UTC (Tue) by mezcalero (subscriber, #45103) [Link]
Ahem. It's not pronounced "pottering".
The 'oe' is actually pronounced like the german Umlaut 'ö', as described on Wikipedia here:
http://en.wikipedia.org/wiki/Open-mid_front_rounded_vowel
That said writing 'Pöttering' would be wrong.
Poettering: Measuring Lock Contention Posted Sep 16, 2009 7:21 UTC (Wed) by michaeljt (subscriber, #39183) [Link]
> Ahem. It's not pronounced "pottering".
I believe (!) that is what the OP was saying in the first place.
Poettering: Measuring Lock Contention Posted Sep 16, 2009 1:28 UTC (Wed) by flewellyn (subscriber, #5047) [Link]
Pronunciation issues aside, I second the motion!
Poettering: Measuring Lock Contention Posted Sep 16, 2009 16:12 UTC (Wed) by Doogie (guest, #59626) [Link] Hoorah, now I can look forward to retiring to a life a poettering around the house. :)
Poettering: Measuring Lock Contention Posted Sep 21, 2009 14:57 UTC (Mon) by jengelh (subscriber, #33263) [Link]
I propose: 1 La = threads waiting on a specific lock × seconds held.
Poettering: Measuring Lock Contention Posted Sep 16, 2009 11:06 UTC (Wed) by ledow (guest, #11753) [Link]
Is it just me, or does "pottering" not suit the measurement in discussion anyway - after all, the computer is just sitting them "pottering" about until the lock is released!
|
|
||||||||||||