Comment #6 on issue 594 by amit.bhayani: Adding cache to AudioPlayer
http://code.google.com/p/mobicents/issues/detail?id=594
Attached is the profile and Jitter graph for with and without cache.
Profiling was done exactly for five min for both scenario and same
environment.
Jitter graph is taken after every 10 Secs and test was run for 3 min at
rate of 1 cps
and announcement length of 10 secs. Hence 10 concurrent announcements
My observations
1) Didn't really find much difference in jitter for with/without cache
2) The absolute time taken by
javax.sound.sampled.AudioSystem.getAudioInputStream
which includes opening file is 2.5x for without cache compared to cache.
Look for
line '0.0% - 33,981 µs javax.sound.sampled.AudioSystem.getAudioInputStream
(line:
358)' with cache and '0.0% - 76,776 µs
javax.sound.sampled.AudioSystem.getAudioInputStream (line: 294)' without
cache.
However we still see jitter. Probably those big jitters are coming from
somewhere else.
In my opinion cache will be useful only if we get rid of Buffer.copy() or
optimize it
to a extent that it consumes very less CPU.
Attachments:
Archive.zip 2.6 MB
--
You received this message because you are listed in the owner
or CC fields of this issue, or because you starred this issue.
You may adjust your issue notification preferences at:
http://code.google.com/hosting/settings
