[asterisk-users] Really Big Queues

[lenz]

Hello Chris,
we have a number of clients who deployed very large CCs over the 200 agent  
range.

Your idea #1 is pretty sound and I believe that's what most people are  
doing. I would like to add a couple of points of attention:
- having hundreds of agents on a box means a lot of synchronous audio  
flowing in and out, so you don't want to save on the ethernet hardware :)
- think about a passive SIP monitoring for call recording, so that you can  
have a different box (or set of boxes) to handle that without slowing down  
the overtaxed ethernet connections of the queueing servers.
I hope this helps,
l.

In data Wed, 17 Jan 2007 02:32:35 +0100, Christopher Snell  
<chris.snell at gmail.com> ha scritto:

> Hi,
>
> How do you folks handle really large queues (350+ simultaneous
> callers) in your Asterisk PBXes?
>
> We're going to be bringing in around 16 PRIs' worth of inbound
> callers, doing skills-based routing, and queuing them up for
> approximately 200 agents.
>
> What's the best way to handle all of these callers?  We want to record
> the calls and we'll probably use the ramdisk method that has been
> discussed on this list.
>
> Here's some ideas that I'm considering:
>
> Idea #1:   Use servers with (2) Digium 4-port PRI cards, running
> Asterisk, as media gateways.  From here, send calls to 2 or more
> Asterisk queue servers.  For each incoming call, run an AGI on the
> media gateways that determines which queue server is least loaded.
> Send this incoming call to the queue server over an IAX2 trunk.  The
> problem with this method is that the queues are not unified; if one
> queue server suddenly has available agents, queued callers on the
> other queue server cannot be (easily?) transfered to the server with
> available agents.  Also, running an AGI for each incoming call is lame
> and slow.
>
> Idea #2:   Use 3com VCX V7122 media gateways to terminate the PRIs and
> send the calls to a load balanced pair of SER proxies.  These proxies
> will somehow keep track of the state of the Asterisk queue servers and
> redirect the incoming calls to the least loaded (most available) queue
> server.  The problem with this method is that, by using SIP, we'll
> probably see higher interrupt load on the Asterisk queue servers.
> Additionally, I'm not a SER expert yet and I have no idea how to get
> SER to monitor the state of the Asterisk queue servers.  As with Idea
> #1, the queues are also not unified, which sucks.
>
> Idea #3:   ???????  (profit!)
>
> Do you fine folks have any ideas or suggestions?
>
> thanks,
>
> Chris



-- 
Home of QueueMetrics - http://queuemetrics.loway.it