Hi List, Why FreeSwitch can handle more then 1,000CC and asterisk only 25CC ? What technology FreeSwitch is used and asterisk don't. I don't know it's the right or wrong but this question come to my mind... -- Thanks and regards Virendra Bhati +91-8885268942 Software Engineer E-mail-: virbhati at gmail.com Skype id:- virbhati2 -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://lists.digium.com/pipermail/asterisk-users/attachments/20120207/22f9f989/attachment.htm>
You mean concurrent calls? You can have several 100 concurrent calls with a good CPU in newer versions of asterisk, however calls per secons (CPS) have some limitations I guess reason being that both are different in Architecture, Asterisk was designed keeping PBX in mind but Freeswitch was for SIP switching Regards, Zohair Raza On Tue, Feb 7, 2012 at 3:38 PM, virendra bhati <virbhati at gmail.com> wrote:> Hi List, > > Why FreeSwitch can handle more then 1,000CC and asterisk only 25CC ? What > technology FreeSwitch is used and asterisk don't. I don't know it's the > right or wrong but this question come to my mind... > > -- > > Thanks and regards > > Virendra Bhati > +91-8885268942 > Software Engineer > E-mail-: virbhati at gmail.com > Skype id:- virbhati2 > > > -- > _____________________________________________________________________ > -- Bandwidth and Colocation Provided by http://www.api-digital.com -- > New to Asterisk? Join us for a live introductory webinar every Thurs: > http://www.asterisk.org/hello > > asterisk-users mailing list > To UNSUBSCRIBE or update options visit: > http://lists.digium.com/mailman/listinfo/asterisk-users >-------------- next part -------------- An HTML attachment was scrubbed... URL: <http://lists.digium.com/pipermail/asterisk-users/attachments/20120207/44e474cc/attachment.htm>
On Tue, 7 Feb 2012 17:08:18 +0530, virendra bhati <virbhati at gmail.com> wrote:>Why FreeSwitch can handle more then 1,000CC and asterisk only 25CC ? What >technology FreeSwitch is used and asterisk don't. I don't know it's the >right or wrong but this question come to my mind...Provided Asterisk, even in release 1.8 or 10, does handle much fewer concurrent calls than Freeswitch, you might find the answer in those articles: "How does FreeSWITCH compare to Asterisk?" www.freeswitch.org/node/117 "Asterisk vs FreeSWITCH" www.richappsconsulting.com/blog/blog-detail/asterisk-vs-freeswitch/ "Asterisk vs. FreeSWITCH" www.anders.com/cms/266 "Open Source VoIP: Asterisk or FreeSwitch?" www.zdnet.com/blog/greenfield/open-source-voip-asterisk-or-freeswitch/233 "FreeSwitch vs Asterisk" www.dslreports.com/forum/r23246683-FreeSwitch-vs-Asterisk
My Asterisk 1.4.19 happily excepts up to 80 concurrent calls (the most I've seen so far) which sends the CPU load up to ~20% on a fairly old server. In our busiest period, from 8 to 8:05 I see up to 200 incoming calls, somewhat less than one call/second. My superiors want to expand and increase the number of clients significantly and the scalability of Asterisk is beginning to worry me. Someone mentioned a "roof" of 250 CC in Asterisk after which stability and call quality becomes increasingly affected. My plan is to implement load-balancing using DUNDi with one extra server initially, and a second available on site for further expansion. This should enable me to accommodate ten times our current load without any significant problems (I hope!), and adding more servers is fairly easy (although I guess there are diminishing returns?). When it comes to the long term I must admit I am increasingly looking at trying out FreeSwitch, the configuration might be trickier but scalability is much higher on my list of priorities. Fra: asterisk-users-bounces at lists.digium.com [mailto:asterisk-users-bounces at lists.digium.com] P? vegne af virendra bhati Sendt: 7. februar 2012 12:38 Til: Asterisk Users Mailing List - Non-Commercial Discussion Emne: [asterisk-users] Asterisk V/s FreeSwitch Hi List, Why FreeSwitch can handle more then 1,000CC and asterisk only 25CC ? What technology FreeSwitch is used and asterisk don't. I don't know it's the right or wrong but this question come to my mind... -- Thanks and regards Virendra Bhati +91-8885268942 Software Engineer E-mail-: virbhati at gmail.com<mailto:virbhati at gmail.com> Skype id:- virbhati2 -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://lists.digium.com/pipermail/asterisk-users/attachments/20120208/24549aa4/attachment.htm>
[This email is either empty or too large to be displayed at this time]
Zohair
What kind of hardware spec are you running CPU, MEM, Drives?
Thanks
Bryant Zimmerman (ZK Tech Inc.)
616-855-1030 Ext. 2003
----------------------------------------
From: "Zohair Raza" <engineerzuhairraza at gmail.com>
Sent: Wednesday, February 08, 2012 3:08 AM
To: "Asterisk Users Mailing List - Non-Commercial Discussion"
<asterisk-users at lists.digium.com>
Subject: Re: [asterisk-users] Asterisk V/s FreeSwitch
Virendra,
You can test your box with sipp
http://etel.wiki.oreilly.com/wiki/index.php/Using_SIPp_to_Stress_Test_Asterisk
I have verified my Asterisk 1.8 box handling 500 concurrent calls and 15 calls
per seconds with 20% cpu, without transcoding.
Regards,
Zohair Raza
On Wed, Feb 8, 2012 at 11:53 AM, Brynjolfur Thorvardsson <binni at
itanet.nu> wrote:
My Asterisk 1.4.19 happily excepts up to 80 concurrent calls (the most
I've seen so far) which sends the CPU load up to ~20% on a fairly old
server. In our busiest period, from 8 to 8:05 I see up to 200 incoming calls,
somewhat less than one call/second. My superiors want to expand and increase
the number of clients significantly and the scalability of Asterisk is beginning
to worry me. Someone mentioned a "roof" of 250 CC in Asterisk after
which stability and call quality becomes increasingly affected. My plan is to
implement load-balancing using DUNDi with one extra server initially, and a
second available on site for further expansion. This should enable me to
accommodate ten times our current load without any significant problems (I
hope!), and adding more servers is fairly easy (although I guess there are
diminishing returns?). When it comes to the long term I must admit I am
increasingly looking at trying out FreeSwitch, the configuration might be
trickier but scalability is much higher on my list of priorities. Fra:
asterisk-users-bounces at lists.digium.com [mailto:asterisk-users-bounces at
lists.digium.com] P? vegne af virendra bhati
Sendt: 7. februar 2012 12:38
Til: Asterisk Users Mailing List - Non-Commercial Discussion
Emne: [asterisk-users] Asterisk V/s FreeSwitch Hi List,
Why FreeSwitch can handle more then 1,000CC and asterisk only 25CC ? What
technology FreeSwitch is used and asterisk don't. I don't know it's
the right or wrong but this question come to my mind...
-- Thanks and regards
Virendra Bhati
+91-8885268942
Software Engineer
E-mail-: virbhati at gmail.com
Skype id:- virbhati2
--
_____________________________________________________________________
-- Bandwidth and Colocation Provided by http://www.api-digital.com --
New to Asterisk? Join us for a live introductory webinar every Thurs:
http://www.asterisk.org/hello
asterisk-users mailing list
To UNSUBSCRIBE or update options visit:
http://lists.digium.com/mailman/listinfo/asterisk-users
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.digium.com/pipermail/asterisk-users/attachments/20120208/bb328921/attachment.htm>
---------------------------------------- From: "Jeff Brower" <jbrower at signalogic.com> Sent: Wednesday, February 08, 2012 8:49 AM To: "Brynjolfur Thorvardsson" <binni at itanet.nu> Subject: Re: [asterisk-users] Asterisk V/s FreeSwitch Brynjolfur-> According to this article here: > > http://anders.com/cms/266 > > the difference mainly lies in how FreeSwitchs handles open > channels in comparison with Asterisk. FS uses one thread > per channel while * keeps jumping between threads. At least > that's how I understand it.If the difference really is 10:1, then I doubt that threads vs. linked lists completely explains it. But the difference may not be that much, as some other posts indicate. I would suggest to Virendra to make sure he's comparing identical configurations: machine type/speed/mem, same type of calls, same amount of call RTP handling (G711, no echo can, no recording, no DTMF, etc), latest versions of both softwares, and so on. That would be a good test. Since the metric in this case is concurrent calls, not CPS, it could be that for some reason, Asterisk's RTP coding isn't as efficient. -Jeff> Fra: asterisk-users-bounces at lists.digium.com [mailto:asterisk-users-bounces at lists.digium.com] P? vegne af virendra > bhati > Sendt: 8. februar 2012 06:34 > Til: Asterisk Users Mailing List - Non-Commercial Discussion > Emne: Re: [asterisk-users] Asterisk V/s FreeSwitch > > thanks Gilles, > > After reading these web links. it's pretty clear that FreeSwitch is batter then Asterisk feature, quality wise. But > asterisk is easy to used. > > But the question is still open from my end. > > How FreeSwitch can support 1000CC but asterisk not ? > > Because FreeSwitch used XML as configuration and asterisk plan text file ? > FreeSwitch used sofia_sip and asterisk used sip ? > Asterisk is PBX and FreeSwitch is SoftSwitch ? > > On Tue, Feb 7, 2012 at 9:10 PM, Gilles <codecomplete at free.fr<mailto:codecomplete at free.fr>> wrote: > On Tue, 7 Feb 2012 17:08:18 +0530, virendra bhati <virbhati at gmail.com<mailto:virbhati at gmail.com>> > wrote: >>Why FreeSwitch can handle more then 1,000CC and asterisk only 25CC ? What >>technology FreeSwitch is used and asterisk don't. I don't know it's the >>right or wrong but this question come to my mind... > Provided Asterisk, even in release 1.8 or 10, does handle much fewer > concurrent calls than Freeswitch, you might find the answer in those > articles: > > "How does FreeSWITCH compare to Asterisk?" > www.freeswitch.org/node/117<http://www.freeswitch.org/node/117> > > "Asterisk vs FreeSWITCH" > www.richappsconsulting.com/blog/blog-detail/asterisk-vs-freeswitch/<http://www.richappsconsulting.com/blog/blog-detail/asterisk-vs-freeswitch/> > > "Asterisk vs. FreeSWITCH" > www.anders.com/cms/266<http://www.anders.com/cms/266> > > "Open Source VoIP: Asterisk or FreeSwitch?" > www.zdnet.com/blog/greenfield/open-source-voip-asterisk-or-freeswitch/233<http://www.zdnet.com/blog/greenfield/open-source-voip-asterisk-or-freeswitch/233> > > "FreeSwitch vs Asterisk" > www.dslreports.com/forum/r23246683-FreeSwitch-vs-Asterisk<http://www.dslreports.com/forum/r23246683-FreeSwitch-vs-Asterisk> > > > > Thanks and regards > > Virendra Bhati > +91-8885268942 > Software Engineer > E-mail-: virbhati at gmail.com<mailto:virbhati at gmail.com> > Skype id:- virbhati2I too am asking which is a better long term bet as my core Asterisk or Freeswitch.. Most of the info offered is from 2008.. The question I am asking is which is better today. In 2008 I was using 1.4 and it was bleeding edge. 1.8 and 10 have gotten a lot more efficent and. i've found I can speed up asterisk if I reduce my call code complexity, If I don't handle media. So my realy question is apples to apples how do the two compair Today? What features do I have to give up with free switch if I want 300 calls per second and 3000 concurrent calls? Then If I strip back asterisk to that feature set how would it stack up? Bryant -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://lists.digium.com/pipermail/asterisk-users/attachments/20120208/de63575e/attachment.htm>
Am 07.02.12 12:38, schrieb virendra bhati:> Hi List, > > Why FreeSwitch can handle more then 1,000CC and asterisk only 25CC ? What > technology FreeSwitch is used and asterisk don't. I don't know it's the > right or wrong but this question come to my mind... >I had done some load tests with asterisk 10 and my highest results was: 1750 calls per seconds up to 13000 concurrent calls done on a intel xeon with dual six core and hyperthreading (= 24 cores) and 12 GB ram. the sysload was around 2.5 during this test. so i am not impressed by 1000 concurrent calls. best regards stefan
---------------------------------------- From: "Stefan Schmidt" <sst at sil.at> Sent: Thursday, February 09, 2012 6:45 AM To: asterisk-users at lists.digium.com Subject: Re: [asterisk-users] Asterisk V/s FreeSwitch Am 07.02.12 12:38, schrieb virendra bhati:> Hi List, > > Why FreeSwitch can handle more then 1,000CC and asterisk only 25CC ?What> technology FreeSwitch is used and asterisk don't. I don't know it's the > right or wrong but this question come to my mind... >I had done some load tests with asterisk 10 and my highest results was: 1750 calls per seconds up to 13000 concurrent calls done on a intel xeon with dual six core and hyperthreading (= 24 cores) and 12 GB ram. the sysload was around 2.5 during this test. so i am not impressed by 1000 concurrent calls. best regards stefan ---------------------------------------------------------------------------- ------------- Stefan This is on target with my configuration I am working on. What kind of dialplan were you using when running the tests. Were you doing database lookups or just answering the calls and playing hold music. Any example would be appreciated so we can quantify your test results. I look forward to your response. Thanks Bryant -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://lists.digium.com/pipermail/asterisk-users/attachments/20120209/7e52d984/attachment.htm>
---------------------------------------- From: "Stefan Schmidt" <sst at sil.at> Sent: Thursday, February 09, 2012 8:24 AM To: asterisk-users at lists.digium.com Subject: Re: [asterisk-users] Asterisk V/s FreeSwitch just done the test again. 13500 concurrent calls at 1750 cps with open rtp ports but without much media transportet, only signaling. see attached screenshot. 10000 concurrent calls with media playing musiconhold but i only have a 100mbit connection on this server so i cant do more here. the version is asterisk 1.8 - unleashed-the-beast which is my own dev branch which has some important performance backports from 10. best regards ps: also this is the output of sipp: sipp -m 13500 -r 1750 -sf sipload.xml -mi 213.x.x.x 213.y.y.y:5060 Resolving remote host '213.x.x.x'... Done. ------------------------------ Scenario Screen -------- [1-9]: Change Screen -- Call-rate(length) Port Total-time Total-calls Remote-host 1750.0(0 ms)/1.000s 5061 22.76 s 13500 213.x.x.x:5060(UDP) Call limit reached (-m 13500), 0.685 s period 0 ms scheduler resolution 0 calls (limit 78750) Peak was 13500 calls, after 7 s 0 Running, 0 Paused, 3 Woken up 0 out-of-call msg (discarded) 1 open sockets Messages Retrans Timeout Unexpected-Msg INVITE ----------> 13500 0 100 <---------- 13500 0 0 180 <---------- 0 0 0 200 <---------- 13500 0 0 ACK ----------> 13500 0 Pause [ 15.0s] 13500 0 BYE ----------> 13500 0 0 200 <---------- 13500 0 0 ------------------------------ Test Terminated -------------------------------- ----------------------------- Statistics Screen ------- [1-9]: Change Screen -- Start Time | 2012-02-09 12:50:06 Last Reset Time | 2012-02-09 12:50:28 Current Time | 2012-02-09 12:50:29 -------------------------+---------------------------+---------------------- ---- Counter Name | Periodic value | Cumulative value -------------------------+---------------------------+---------------------- ---- Elapsed Time | 00:00:00:684 | 00:00:22:763 Call Rate | 0.000 cps | 593.068 cps -------------------------+---------------------------+---------------------- ---- Incoming call created | 0 | 0 OutGoing call created | 0 | 13500 Total Call created | | 13500 Current Call | 0 | -------------------------+---------------------------+---------------------- ---- Successful call | 1174 | 13500 Failed call | 0 | 0 -------------------------+---------------------------+---------------------- ---- Call Length | 00:00:15:015 | 00:00:15:011 ------------------------------ Test Terminated -------------------------------- Am 09.02.12 12:57, schrieb Sammy Govind:> Wow, > I bet even asterisk developers wouldn't believe so. What have they done!.> No, actually can you tell if server was processing media along with the > calls as well !? > > I once tested without media and really I had some 1000+ CCs on asterisk > server on a regular dev machine with choppy audio on an actual callwhile> still under stress. > > Kindly please confirm your stats. > > Regards, > Sammy > > On Thu, Feb 9, 2012 at 4:49 PM, Stefan Schmidt <sst at sil.at> wrote: > >> Am 07.02.12 12:38, schrieb virendra bhati: >>> Hi List, >>> >>> Why FreeSwitch can handle more then 1,000CC and asterisk only 25CC ?What>>> technology FreeSwitch is used and asterisk don't. I don't know it'sthe>>> right or wrong but this question come to my mind... >>> >> I had done some load tests with asterisk 10 and my highest results was: >> >> 1750 calls per seconds up to >> 13000 concurrent calls >> >> done on a intel xeon with dual six core and hyperthreading (= 24 cores) >> and 12 GB ram. the sysload was around 2.5 during this test. >> >> so i am not impressed by 1000 concurrent calls. >> >> best regards >> >> stefan >> >> --Stefan Did you use a published guide to setup your test or would you be willing to give some kind of step by step so I could setup a Similar test I would like to see how far I can push several of our builds. I really would like to get solid numbers on some of my configs. Thanks Bryant -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://lists.digium.com/pipermail/asterisk-users/attachments/20120209/65ade634/attachment.htm>
Am 09.02.12 14:19, schrieb Bryant Zimmerman:> Stefan > > This is on target with my configuration I am working on. What kind of > dialplan were you using when running the tests. > Were you doing database lookups or just answering the calls and playing > hold music. Any example would be appreciated so we can quantify your test > results. I look forward to your response. > > Thanks > Bryantthe dialplan is quite simple: for the signaling up to 13500 CC i use this wait and for the 10000CC i enable the musiconhold exten => monitor,1,Noop(PERFORMANCE TESTS) exten => monitor,n,Answer ;exten => monitor,n,MusicOnHold(806,45) exten => monitor,n,Wait(45) exten => monitor,n,Hangup and i have attached the sipp scenario i was using which is also very simple. as i said i only have a 100 mbit connection on this server and its also only a virtual machine and on this physical host there are also some production machines running so i cant put it even further. very important is the rtp.conf where you have to change the rtpend port range up to 65000 or you wouldnt be able to open enough rtp ports. just to point some things out: asterisk below 1.8 was getting even slower with every version. before asterisk 1.8 was released i allready wrote some performance patches which made sip handling better and faster but the big performance boosts came too late for 1.8 but they are in asterisk 10 and the performance increase in sip udp handling is around 2 to 3 times faster. btw my normal production machines which are just the same virtual machines like this test system. i also had 330 concurrent calls, some with transcoding, many database lookups, musiconhold, pickup ... and the sysload was around 1.0 ;) best regards stefan -------------- next part -------------- A non-text attachment was scrubbed... Name: sipload.xml Type: text/xml Size: 1742 bytes Desc: not available URL: <http://lists.digium.com/pipermail/asterisk-users/attachments/20120209/9512ab21/attachment.bin>
On 09-02-12 14:52, Stefan Schmidt wrote:> Am 09.02.12 14:19, schrieb Bryant Zimmerman: >> Stefan >> >> This is on target with my configuration I am working on. What kind of >> dialplan were you using when running the tests. >> Were you doing database lookups or just answering the calls and playing >> hold music. Any example would be appreciated so we can quantify your test >> results. I look forward to your response. >> >> Thanks >> Bryant > > the dialplan is quite simple: > > for the signaling up to 13500 CC i use this wait and for the 10000CC i > enable the musiconhold > > exten => monitor,1,Noop(PERFORMANCE TESTS) > exten => monitor,n,Answer > ;exten => monitor,n,MusicOnHold(806,45) > exten => monitor,n,Wait(45) > exten => monitor,n,HangupIirc a long time ago there was a discussion about load testing by playing MoH was not a realistic test. Something about all MoH music getting streamed synchronized so basically Asterisk only has to stream one file and sorta multiplex that single output to all the established calls (legs). [snip]> btw my normal production machines which are just the same virtual > machines like this test system. i also had 330 concurrent calls, some > with transcoding, many database lookups, musiconhold, pickup ... and the > sysload was around 1.0 ;)The difference (13500 with MoH versus 330 with a real dialplan) shows that it makes sense to mimic your dialplan in your test scenario as much as possible to see how far you can realistically push the box and still keep things stable and sound quality good. Regards, Patrick
---------------------------------------- From: "Patrick Lists" <asterisk-list at puzzled.xs4all.nl> Sent: Thursday, February 09, 2012 10:42 AM To: asterisk-users at lists.digium.com Subject: Re: [asterisk-users] Asterisk V/s FreeSwitch On 09-02-12 14:52, Stefan Schmidt wrote:> Am 09.02.12 14:19, schrieb Bryant Zimmerman: >> Stefan >> >> This is on target with my configuration I am working on. What kind of >> dialplan were you using when running the tests. >> Were you doing database lookups or just answering the calls and playing >> hold music. Any example would be appreciated so we can quantify yourtest>> results. I look forward to your response. >> >> Thanks >> Bryant > > the dialplan is quite simple: > > for the signaling up to 13500 CC i use this wait and for the 10000CC i > enable the musiconhold > > exten => monitor,1,Noop(PERFORMANCE TESTS) > exten => monitor,n,Answer > ;exten => monitor,n,MusicOnHold(806,45) > exten => monitor,n,Wait(45) > exten => monitor,n,HangupIirc a long time ago there was a discussion about load testing by playing MoH was not a realistic test. Something about all MoH music getting streamed synchronized so basically Asterisk only has to stream one file and sorta multiplex that single output to all the established calls (legs). [snip]> btw my normal production machines which are just the same virtual > machines like this test system. i also had 330 concurrent calls, some > with transcoding, many database lookups, musiconhold, pickup ... and the > sysload was around 1.0 ;)The difference (13500 with MoH versus 330 with a real dialplan) shows that it makes sense to mimic your dialplan in your test scenario as much as possible to see how far you can realistically push the box and still keep things stable and sound quality good. Regards, Patrick ------------------------------------- Patrick I agree with you but it looks like these test show that asterisk 10 could handle a very high volume switch style application if you remove the rtp and media handling from the dialplan. For me this opens up the question as to why would I need freeswitch for high volume switching if my RTP is being handled else where. I would have a different dialplan code set for this kind of application anyway. This is why I asked Stefan to share his dialplan and testing matrix. It allows us to see raw performance what is the asterisk code base cable of and for his test case it is quite impressive. For other test cases it may not be. I would like to see someone test a full PBX implementation with independent audio files playing not just MOH and see what kind of load it can handle there. That would give me a better idea of how the two would compare. I wounder how the new 10.0 conference application would stack against freeswitches conference. How would we best design a test for that? This is a great conversation. Very productive for me at least. Thanks Bryant -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://lists.digium.com/pipermail/asterisk-users/attachments/20120209/fd44de4b/attachment.htm>
Am 09.02.12 16:45, schrieb Patrick Lists:> Iirc a long time ago there was a discussion about load testing by > playing MoH was not a realistic test. Something about all MoH music > getting streamed synchronized so basically Asterisk only has to stream > one file and sorta multiplex that single output to all the established > calls (legs).this load tests are mostly about sip signal handling and not so much about rtp streaming but this moh class which i use had 100 files and random set to yes, so its atleast not soo bad.> [snip] > >> btw my normal production machines which are just the same virtual >> machines like this test system. i also had 330 concurrent calls, some >> with transcoding, many database lookups, musiconhold, pickup ... and the >> sysload was around 1.0 ;) > > The difference (13500 with MoH versus 330 with a real dialplan) shows > that it makes sense to mimic your dialplan in your test scenario as much > as possible to see how far you can realistically push the box and still > keep things stable and sound quality good.This 330 concurrent calls was only the highest value which i had on a normal production system and its really hard to build a test setup which presents a system with 4000 sip peer doing some calls. but the sound quality was still good even with 10000 calls in my tests.> Regards, > Patrick >best regards stefan
If the MOH thing is really true, a more "realistic" test would be to run playback(demo-instruct). Since I know that I will eventually cross this bridge in real life/real time, I devised this test on my Asterisk 10.0 box Dialplan (in default context) exten => 3366,1,answer() exten => 3366,n,playback(demo-instruct,noanswer) exten => 3366,n,playback(demo-instruct,noanswer) exten => 3366,n,playback(vm-goodbye,noanswer) exten => 3366,n,hangup() SIPP command ./sipp -l 399 -d 99000 -m 399 -s 3366 -p 5061 -sn uac 127.0.0.1 -trace_err I was able to do 260 concurrent calls with no issues. The 2 playbacks for demo-instruct were to cover 99 seconds since the file is only 67 seconds long. For the 300/1000 call scenario, you would need to duplicate the line accordingly. The limiting factor for me was my rtp.conf. I set up a range of 10001-10520 which stopped at 260 since each "call" allocates 4 rtp slots (2 in use and 2 for transfer, etc). -----Original Message----- From: asterisk-users-bounces at lists.digium.com [mailto:asterisk-users-bounces at lists.digium.com] On Behalf Of Stefan Schmidt Sent: Thursday, February 09, 2012 10:06 AM To: asterisk-users at lists.digium.com Subject: Re: [asterisk-users] Asterisk V/s FreeSwitch Am 09.02.12 16:45, schrieb Patrick Lists:> Iirc a long time ago there was a discussion about load testing by > playing MoH was not a realistic test. Something about all MoH music > getting streamed synchronized so basically Asterisk only has to stream > one file and sorta multiplex that single output to all the established > calls (legs).this load tests are mostly about sip signal handling and not so much about rtp streaming but this moh class which i use had 100 files and random set to yes, so its atleast not soo bad.> [snip] > >> btw my normal production machines which are just the same virtual >> machines like this test system. i also had 330 concurrent calls, some >> with transcoding, many database lookups, musiconhold, pickup ... and >> the sysload was around 1.0 ;) > > The difference (13500 with MoH versus 330 with a real dialplan) shows > that it makes sense to mimic your dialplan in your test scenario as > much as possible to see how far you can realistically push the box and > still keep things stable and sound quality good.This 330 concurrent calls was only the highest value which i had on a normal production system and its really hard to build a test setup which presents a system with 4000 sip peer doing some calls. but the sound quality was still good even with 10000 calls in my tests.> Regards, > Patrick >best regards stefan -- _____________________________________________________________________ -- Bandwidth and Colocation Provided by http://www.api-digital.com -- New to Asterisk? Join us for a live introductory webinar every Thurs: http://www.asterisk.org/hello asterisk-users mailing list To UNSUBSCRIBE or update options visit: http://lists.digium.com/mailman/listinfo/asterisk-users