Currently only 5 conferences and 16 channels already have 70% CPU time. Does that seem to be normal?
Do you know any tuning preferences for OPUS codec? Does disabling zrtp help?

I am using FS GIT Master.

Conference 5052916 (4 members rate: 48000 flags: running|answered|enforce_min|dynamic|exit_sound|enter_sound)

Conference 7020910 (1 member rate: 48000 flags: running|answered|enforce_min|dynamic|exit_sound|enter_sound)

Conference 6568068 (1 member rate: 48000 flags: running|answered|enforce_min|dynamic|exit_sound|enter_sound)

Conference 6568069 (6 member rate: 48000 flags: running|answered|enforce_min|dynamic|exit_sound|enter_sound)

Conference 6131817 (4 members rate: 48000 flags: running|answered|enforce_min|dynamic|exit_sound|enter_sound)


top - 12:36:25 up 3 days, 11:33,  4 users,  load average: 0.37, 0.30, 0.31

Tasks: 114 total,   1 running, 113 sleeping,   0 stopped,   0 zombie

%Cpu0  : 24.5 us,  1.5 sy,  0.0 ni, 71.6 id,  0.0 wa,  0.0 hi,  2.4 si,  0.0 st

%Cpu1  :  0.0 us,  0.0 sy,  0.0 ni,100.0 id,  0.0 wa,  0.0 hi,  0.0 si,  0.0 st

%Cpu2  : 17.7 us,  0.0 sy,  0.0 ni, 81.7 id,  0.6 wa,  0.0 hi,  0.0 si,  0.0 st

%Cpu3  :  0.0 us,  0.0 sy,  0.0 ni,100.0 id,  0.0 wa,  0.0 hi,  0.0 si,  0.0 st

%Cpu4  : 32.8 us,  0.0 sy,  0.0 ni, 67.2 id,  0.0 wa,  0.0 hi,  0.0 si,  0.0 st

%Cpu5  : 17.3 us,  0.0 sy,  0.0 ni, 82.7 id,  0.0 wa,  0.0 hi,  0.0 si,  0.0 st

%Cpu6  :  0.0 us,  0.4 sy,  0.0 ni, 99.6 id,  0.0 wa,  0.0 hi,  0.0 si,  0.0 st

%Cpu7  :  0.0 us,  0.0 sy,  0.0 ni,100.0 id,  0.0 wa,  0.0 hi,  0.0 si,  0.0 st

KiB Mem:  32710216 total,  1451804 used, 31258412 free,   154716 buffers

KiB Swap: 16768892 total,        0 used, 16768892 free,   969216 cached


  PID USER      PR  NI  VIRT  RES  SHR S  %CPU %MEM    TIME+  COMMAND                                                                                                                                       

32470 root      -2 -10  501m  98m 9912 S  71.2  0.3 156:50.85 freeswitch



--- Ursprüngliche Nachricht ---
Von: Michael Jerris <mike@jerris.com>
Datum: 27.05.2014 15:46:03
An: FreeSWITCH Users Help <freeswitch-users@lists.freeswitch.org>
Betreff: Re: [Freeswitch-users] scalability of mod_conference and mod_shout

I wouldn't want to make guesses at numbers, but cpu will be your limit, and codecs are by far the biggest contributor to that cost, you will get many more participants with ulaw than you do with opus.  I would experiment using real talking patterns as this will effect results.

Mike

On May 25, 2014, at 10:24 PM, <martin.hoole@emailn.de> <martin.hoole@emailn.de> wrote:

I'm new to FreeSWITCH and I'm trying to find more information on the scalability of mod_conference and mod_shout.
I'am aware that stress testing should be better done by some professional consulting from freeswitch team @ Anthony, Michael, Brian.

But before further investigating/spending money in freeswitch I would be glad to get some infos if FS is the right solution to accomplish our use case:

On a Intel Core i7-3770 with 32 GB Ram can you tell me some estimated values about 

- maximum concurrent conferences with average 3 speakers and 10 listeners
- publishing each conference to icecast with mod_shout

Does the codec has an big impact in terms of concurrent connection if using OPUS or PCM and also mod_shout (transcoding)?

I would appreciate a rough estimate about your experience values.

Cheers,

martin






Versendet mit Emailn.de - Freemail

* Unbegrenzt Speicherplatz
* Eigenes Online-Büro
* 24h besten Mailempfang
* Spamschutz, Adressbuch