[Freeswitch-dev] Scalabilty of Freeswitch

Thu Aug 27 22:55:20 PDT 2009

I would like to comment on this statement:

“3) 9.9/10 newcomers who try to do load testing get it wrong somewhere and we spend a month dealing with them on the list etc.”

 I see this very often in this mailing-list. Would it be a good idea to create wiki page how to fine tune Freeswitch to handle as many calls as possible?

Maybe such page already exist? (I’m not able to find it). If not – I would like to help creating it. 

After few months when we will start developing our new product – I will come here and will cry for help (I will be one of these 9.9/10) to tune FS to handle as many calls as possible, because that will be my goal.

Regards,

Mindaugas Kezys

http://www.kolmisoft.com

VoIP Billing and Routing Solutions

From: freeswitch-dev-bounces at lists.freeswitch.org [mailto:freeswitch-dev-bounces at lists.freeswitch.org] On Behalf Of Anthony Minessale
Sent: 2009 m. rugpjūčio 27 d. 20:07
To: freeswitch-dev at lists.freeswitch.org
Subject: Re: [Freeswitch-dev] Scalabilty of Freeswitch

I think there are 3 main problems.

1) The core developers should not be blowing their own horn about performance results in any official capacity because it will appear skewed to skeptics.
2) The core developers are too busy to bother with gathering the stats.
3) 9.9/10 newcomers who try to do load testing get it wrong somewhere and we spend a month dealing with them on the list etc.

It sort of like if someone offered FREE CARS, take one and drive away and do whatever you want with it.
People show up and say, "how fast can it go?" "how well does it handle on curves?" .. I don't know it's a FREE CAR, go drive it and you tell me.

I think some people forget this is an open source project and it's up to the community to decide for themselves how well it works.

I would love to find a way to save us all the time of dealing with people who try to load test first and try real calls later.

The industry standard for calls is 50cps which is assuming a typical load of 25cps burst-able to 50 to cover a fault in a cluster.

Most people doing 50cps all day long should be making so much money they can either afford nicer boxes or more expensive equipment if
they need to scale more.

FS can handle this load easily and many people tune their box to do triple that if not even more but that is all subjective to their
hardware choices etc.

sigh,

On Thu, Aug 27, 2009 at 11:32 AM, Michael Giagnocavo <mgg at giagnocavo.net> wrote:

Can’t we do what some vendors do? Pick the simplest config for a simple scenario, like statically bridging two channels, then publish those numbers? Or say, use one of the standard SIPP scenarios with no RTP? And so on. 

From: freeswitch-dev-bounces at lists.freeswitch.org [mailto:freeswitch-dev-bounces at lists.freeswitch.org] On Behalf Of Michael Jerris
Sent: Thursday, August 27, 2009 8:44 AM

To: freeswitch-dev at lists.freeswitch.org
Subject: Re: [Freeswitch-dev] Scalabilty of Freeswitch

if you have the time and ability to create such a multi-dementional array of tjat data in some usable form that would be fine, it is a wiki after all.  I however tend to think that there are too many variables to reliably provide any sort of real data and that the time required to do so would be quite a lot.  

Mike

On Aug 27, 2009, at 1:37 AM, Mindaugas Kezys wrote:

Maybe it would be a good idea to create wiki page just to put such kind of information in a table:

That way interested persons could get a grasp what is really all about.

This is very common question based on which many people measure switch capabilities, so in my opinion should be treated with that in mind (as marketing oportunity)

Regards,

Mindaugas Kezys

http://www.kolmisoft.com

VoIP Billing and Routing Solutions

From: freeswitch-dev-bounces at lists.freeswitch.org [mailto:freeswitch-dev-bounces at lists.freeswitch.org] On Behalf Of Michael Collins
Sent: 2009 m. rugpjūčio 26 d. 23:36
To: freeswitch-dev at lists.freeswitch.org
Subject: Re: [Freeswitch-dev] Scalabilty of Freeswitch

This question sounds eerily familiar...

On Wed, Aug 26, 2009 at 1:15 PM, Jerry Richards <jerry.richards at teotech.com> wrote:

Hello All,

Does anyone know what the capacity of a stand-alone Freeswitch (assuming a
top-of-the-line server), in terms of how many users?

You can have many hundreds of users, but there are a lot of factors: network infrastructure, call volume, etc. 

Also, when that number is exceeded, how can Freeswitch server be distributed
to accommodate a larger installation?

Yes there are strategies. You definitely want a professional to assist if this is a serious production environment. There are members of the FS community who do this sort of thing, or you could email consulting at freeswitch.org to get assistance from the core FS developers.
-MC

_______________________________________________
FreeSWITCH-dev mailing list
FreeSWITCH-dev at lists.freeswitch.org
http://lists.freeswitch.org/mailman/listinfo/freeswitch-dev
UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-dev
http://www.freeswitch.org

_______________________________________________
FreeSWITCH-dev mailing list
FreeSWITCH-dev at lists.freeswitch.org
http://lists.freeswitch.org/mailman/listinfo/freeswitch-dev
UNSUBSCRIBE:http://lists.freeswitch.org/mailman/options/freeswitch-dev
http://www.freeswitch.org

-- 
Anthony Minessale II

FreeSWITCH http://www.freeswitch.org/
ClueCon http://www.cluecon.com/
Twitter: http://twitter.com/FreeSWITCH_wire

AIM: anthm
MSN:anthony_minessale at hotmail.com <mailto:MSN%3Aanthony_minessale at hotmail.com> 
GTALK/JABBER/PAYPAL:anthony.minessale at gmail.com <mailto:PAYPAL%3Aanthony.minessale at gmail.com> 
IRC: irc.freenode.net #freeswitch

FreeSWITCH Developer Conference
sip:888 at conference.freeswitch.org <mailto:sip%3A888 at conference.freeswitch.org> 
iax:guest at conference.freeswitch.org/888
googletalk:conf+888 at conference.freeswitch.org <mailto:googletalk%3Aconf%2B888 at conference.freeswitch.org> 
pstn:213-799-1400

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.freeswitch.org/pipermail/freeswitch-dev/attachments/20090828/10191833/attachment-0001.html 