How can Asterix PBX/IP be compared to vendors such as Avalla, Nortel.

There is virtually no limit to the number of users that can be supported via Asterisk. That said, Asterisk's scalability is, beyond a certain point, cumbersome. For organizations with the expertise to deal with these issues, this is a trivial matter, more than offset by the dramatic cost savings of this open-source solution.

Voice quality is dependent on a number of factors, but all of those being equal, Asterisk will deliver voice quality as good as any other system.

I'm not sure you can make a direct comparison at all, since Asterisk is just software.

miguel

Isn't that how the whole industry is going though? Cisco CallManager/Unity is purely software too.

Personnel costs to deal with it may well be a hell of a lot more expensive than commercial solutions. An oft-overlooked point when people start blathering on about open source.

Asterisk is a software based solutions require skill in set up and maintanence. But the PBXs are made like commodity and needs minimum skill and expertise to set up and maintanence. As mentioned opensource require high tech support and it is costly. But Asterisk offer features and facility far ahead of any PBX. The scalability, quality etc,, are very good for Asterisk and only you need to add more hardware resources. The development in Asterisk going on for embedded linux etc to make the system simple.

What part of "For organizations with the expertise to deal with these issues..." did you not get?

You are making a number of assumptions, after the fact, to justify your position. Given those assumptions, your argument is valid. Nevertheless, it remains true that, given staff of adequate skills, Asterisk can be deployed in the PBX role at substantial savings over traditional hardware.

Indeed, for the money I'd save over an Avaya system with similar capabilites, I could train my PBX-boy or network admin to expert level on Asterisk, or hire a consultant to do the work, and still have tens of thousands of dollars left over.

"Just works" is my yardstick as well, when it comes to systems that are as vital as telephony is to most organizations. Asterisk reached that point some time ago. Set up on proper hardware, by someone who knows what she's doing, it is easily the equal of anthing the traditional PBX vendors have to offer, and exceeds them in many areas.

I didn't miss the point at all. The level of experience needed to run a PBX is quite a bit less than that needed to put up with the same thing cobbled up on a linux box. All too often those interested in preaching about the 'savings' of things like open source fail to appreciate the entire range of actual costs to the organization. Yes, a place that already has a PBX expert on staff might well also be able to handle the added burden of cobbling up a linux box. But I'd argue that's not common enough to make it a better deal for most companies looking for a brainless-to-operate PBX that just 'works'.

When he said "expertise in place", I think he meant someone who is both Linux admin and PBX admin experienced. Which is likely to be rare.

Well Avaya isn't exactly known for dollar value. Compare it against a vendor with more reasonably priced stuff and your argument does not hold up.

this depends on scale and complexity - simple setups are meant to be easy, but like most other IT related systems, things get complicated and need more care and attention as they get bigger.

i think this may actually be a key point.

FWIW - the "expensive" commercial systems such as Cisco call manager tend to spend a big chunk of the total systems cost on end points and gateways. So IP phones, or convertors to support existing analog handsets, and gateways to access PSTN etc is where a lot of money ends up.

The high end softswitches used by PSTN providers are even more biased towards peripheral costs as the individual modes scale up the number of end points per system.

I cant see how having a "free" central call management piece can affect the direct cost of the bits that use separate hardware to scale up.

again on call manager you can put gateways cards into the central processor (or you could initially) - but this severely restricts the number of end points, call setup rates and other scale limits.

to answer the original Q - last time i checked call manager could handle

i cant see any reason that asterix cant go to similar scale with a similar design- but if it did i would want a 2 or 4 hour fix support + maint contract on it, design that would survive loss of any server etc - and i think that might be hard to find for a non commercial product.

One of the advantages with something commercial like the cisco is that they will provide a reference design, and commit to scaling rules, size limits and so on.