[asterisk-biz] Asterisk Performance Results
Zoa
zoachien at securax.org
Fri Nov 16 19:52:39 CST 2007
I would like to comment on my own post, unlike signate before in 2006 ,
according to their pdf they did do their homework and used a setting new
to sipp: :make pcapplay”instead of the echo mode. which should fix that,
and reinvite is set to no, so i cannot find any flaws in the test setup,
however i would still like to see pps measurements, i still can't
believe linux can handle that amount of packets per second in userlevel
on that type of hardware. (based on my own measurements from years ago
on a dual xeon - not dual core)
Zoa
Zoa wrote:
> When doing the packets, can you also monitor the bandwitdh used ? As i
> pointed out before, SIPp does not generate audio and asterisk will not
> send audio packets unless packets are received.
> (This does conflict with the fact that you see higher load when
> transcoding is used).
>
> 400 transcodings is about what i would expect for the cpu's, but the
> 1500 simultaneous calls is more then i would expect from linux user land
> packet handling on that system.
> Did you do any tweaks with the kernel or cpu affinity for this ? What
> network card are you using ? On top load, how much is going to system ?
> What kernel are you using ?
>
> Zoa
>
> Matthew Rubenstein wrote:
>
>> I see from your webpage report that you used SIPp clients to generate
>> the call legs.
>>
>>
>> On Fri, 2007-11-16 at 15:56 -0500, Matthew Rubenstein wrote:
>>
>>
>>> That is extremely valuable baseline data.
>>>
>>> What did you use to generate the test call legs? Would you consider
>>> rerunning the tests, throwing all the call legs into MeetMe conferences
>>> (2 or 4 way) instead of simple call completion, to benchmark
>>> conferencing's capacity requirements vs the baseline? With that info,
>>> most of the feature capacity requirements can be derived to a
>>> per-feature, per-leg basis for capacity planning. Only audio recording
>>> would be left as a major feature, but that's not as common.
>>>
>>>
>>> On Fri, 2007-11-16 at 15:41 -0500, Jim Dalton wrote:
>>>
>>>
>>>> SIP
>>>>
>>>> Jim Dalton
>>>> VoIP Routing, Accounting, Security
>>>> 1.404.526.6053
>>>> www.TransNexus.com
>>>>
>>>>
>>>>
>>>>
>>>>> -----Original Message-----
>>>>> From: Matthew Rubenstein [mailto:email at mattruby.com]
>>>>> Sent: Friday, November 16, 2007 3:35 PM
>>>>> To: Jim Dalton
>>>>> Cc: Asterisk -Biz
>>>>> Subject: RE: [asterisk-biz] Asterisk Performance Results
>>>>>
>>>>> SIP, ZAP or ?
>>>>>
>>>>>
>>>>> On Fri, 2007-11-16 at 15:34 -0500, Jim Dalton wrote:
>>>>>
>>>>>
>>>>>> Each call has two legs: one call leg inbound to the B2BUA
>>>>>>
>>>>>>
>>>>> and one call
>>>>>
>>>>>
>>>>>> leg outbound from the B2BUA.
>>>>>>
>>>>>> 1500/400 calls
>>>>>> 3000/800 call legs
>>>>>>
>>>>>> Jim Dalton
>>>>>> VoIP Routing, Accounting, Security
>>>>>> 1.404.526.6053
>>>>>> www.TransNexus.com
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>>> -----Original Message-----
>>>>>>> From: asterisk-biz-bounces at lists.digium.com
>>>>>>> [mailto:asterisk-biz-bounces at lists.digium.com] On Behalf
>>>>>>>
>>>>>>>
>>>>> Of Matthew
>>>>>
>>>>>
>>>>>>> Rubenstein
>>>>>>> Sent: Friday, November 16, 2007 3:10 PM
>>>>>>> To: Jim Dalton
>>>>>>> Cc: Asterisk -Biz
>>>>>>> Subject: Re: [asterisk-biz] Asterisk Performance Results
>>>>>>>
>>>>>>> Thanks for the data. Are those 1500/400 calls 2-leg calls, or
>>>>>>> individual legs of calls (so really 750/200 2-leg calls)?
>>>>>>>
>>>>>>>
>>>>>>> On Fri, 2007-11-16 at 15:02 -0500, Jim Dalton wrote:
>>>>>>>
>>>>>>>
>>>>>>>> We recently performed an indepth performance test on
>>>>>>>>
>>>>>>>>
>>>>>>> Asterisk V1.4.11
>>>>>>>
>>>>>>>
>>>>>>>> configured as a SIP B2BUA.
>>>>>>>>
>>>>>>>> Asterisk was running on a server with two Xeon 5140, dual
>>>>>>>>
>>>>>>>>
>>>>>>> core, 2.33
>>>>>>>
>>>>>>>
>>>>>>>> GHz CPUs and 4 GB of RAM.
>>>>>>>>
>>>>>>>> We found that an Asterisk B2BUA on this hardware can
>>>>>>>>
>>>>>>>>
>>>>> manage 1500
>>>>>
>>>>>
>>>>>>>> simultaneous calls with no transcoding and 400 simultaneous
>>>>>>>>
>>>>>>>>
>>>>>>> calls with
>>>>>>>
>>>>>>>
>>>>>>>> G.711 to G.729 transcoding.
>>>>>>>>
>>>>>>>> A summary of the test is available at
>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>> http://www.transnexus.com/White%20Papers/asterisk_V1-4-11_performance.
>>>>>
>>>>>
>>>>>>>> htm
>>>>>>>> The test details are available at
>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>> http://www.transnexus.com/White%20Papers/Asterisk_Performance_as_a_S
>>>>>
>>>>>
>>>>>>> IP
>>>>>>>
>>>>>>>
>>>>>>>> _B2BUA
>>>>>>>> .pdf
>>>>>>>>
>>>>>>>> Jim Dalton
>>>>>>>> www.transnexus.com
>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>>>
>>>>>>>> _______________________________________________
>>>>>>>> --Bandwidth and Colocation Provided by
>>>>>>>> http://www.api-digital.com--
>>>>>>>>
>>>>>>>> asterisk-biz mailing list
>>>>>>>> To UNSUBSCRIBE or update options visit:
>>>>>>>> http://lists.digium.com/mailman/listinfo/asterisk-biz
>>>>>>>>
>>>>>>>>
>>>>>>> --
>>>>>>>
>>>>>>> (C) Matthew Rubenstein
>>>>>>>
>>>>>>>
>>>>>>> _______________________________________________
>>>>>>> --Bandwidth and Colocation Provided by
>>>>>>>
>>>>>>>
>>>>> http://www.api-digital.com--
>>>>>
>>>>>
>>>>>>> asterisk-biz mailing list
>>>>>>> To UNSUBSCRIBE or update options visit:
>>>>>>> http://lists.digium.com/mailman/listinfo/asterisk-biz
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>
>>>>>>
>>>>> --
>>>>>
>>>>> (C) Matthew Rubenstein
>>>>>
>>>>>
>>>>>
>>>>>
>>>>>
>>>>
>>>>
>
>
> _______________________________________________
> --Bandwidth and Colocation Provided by http://www.api-digital.com--
>
> asterisk-biz mailing list
> To UNSUBSCRIBE or update options visit:
> http://lists.digium.com/mailman/listinfo/asterisk-biz
>
More information about the asterisk-biz
mailing list