|
Thread Index RE: [isp-clec] VoIP Feature Servers (Asterisk bashing)
[apologies for top-posting, seems this thread is already broken] All - There are systems that are and have been running hundreds of thousands of calls (or "events" since the term "call" is starting to lose it's meaning) per day through Asterisk servers or server clusters, creating large revenues and with acceptable quality to the customers. If you keep your requirements simple and segmented, and you REALLY know what you're doing, Asterisk is a very cost-effective solution for many VoIP tasks. It is not an SS7 gateway (though third-party products exist for this) and lacks support for many "industry" standards which otherwise would be selling points of commercial software if you're acronym-inclined. There are TDM interface cards which work well with Asterisk, but if you're using it for terminating T3's worth of TDM, then you probably should be looking elsewhere for IP-to-TDM gear (at least for the simple conversion process.) Many people are running 8-12 T1's worth in single chassis configurations with decent results, so sizing is all a matter of what the individual circumstances are for investment, abilities, and localized criteria. Outside of the United States, Asterisk is becoming the de-facto standard for newer carriers, at least from my biased view judging by the questions on the Asterisk mailing lists and that which I hear at various conferences on the topic of carrier use of the software. As a feature server it also works very well. You may be using Asterisk in some form or another if you utilize voicemail from some of the larger VoIP carriers whose names would sound familiar to you. As a VoIP router in many situations, it works similarly well. Calling it "broken by design" is misleading and inflammatory, and the assertion that forks are a sign of failure is like saying that Linux must be fundamentally unsound since there are so many different varieties of distributions. It is my belief that Asterisk forks exist due to differences of opinion on how to perform the same task, and also license issues with Digium's version of the GPL which includes a commercial disclaimer. This is the same for any open-source project with such a huge developer community who has commercial interest in the code that drives them to seek perfection in their own ways even if it differs from the group. That is encouraged, and will serve to make Asterisk a more polished software platform as those changes are woven in over time. Additionally, this is the "isp-clec" list, so I suspect that Brian's assertion that "these are carriers here" is probably not applicable to the bulk of the list members - it seems that these are ISPs and IP data network people who have been moving into the voice world, not the other way around. They understand non-certified gear, unwarranted software, and piecemeal solutions since that's the only way they've remained alive for the past ten years. I recall installing disguised "pirate" non-NEBS compliant equipment in telco facilities quite some years back from a startup company called Cisco... Brian B is correct at least in that there is no name-brand support for these type of operations, save Digium. Asterisk is lacking a significant VAR channel due to it's (mostly) open-source nature, and also has a long-lamented (by me) problem of the very high-end users keeping their mouths shut and their patches to themselves. There is a significant developer community who has invested more time and effort into Asterisk than possibly any of the commercial solutions currently deployed, and the open-source model seems to be working well in this case. QC is still questionable as feature sets expand, but is getting better with the "stable" releases. If you are looking for an out-of-the-box solution, stay away from Asterisk. If you need a GUI, a nationwide 24/7 support organization, or don't have someone who can write C code on staff, then you should remove Asterisk from your telco-level deployment option list. However, if you were an early Linux embracer, and you like to be able to be flexible with your abilities on inexpensive hardware, then Asterisk is for you and can bring excellent return on investment with the appropriate planning. Any organization who is considering operating a VoIP feature server of some type would be remiss to not review Asterisk for themselves and ignore both my applause and Brian B's contempt. Test it for yourself and see if it fits your business and product requirements. PS: I'll pick my wording very carefully here to revise Peter's comment on my behalf: "There are no single backplane servers running Asterisk currently to my knowledge that have been rigorously tested with 30,000 or more simultaneous channels of mixed SIP signalling and RTP data, but that is well within the possible scope of a small set of Asterisk platforms designed for the task." Currently, NDAs prevent me from being more inclusive in that statement, but some surfing around with Google or the mailing list archives will find lots of small companies who are doing unusual things with Asterisk scaling. JT Actually, it has been demonstrated to scale to 30k connections by John Todd. -----Original Message----- From: Brian F. G. Bidulock [mailto:bidulock@openss7.org] Sent: Wednesday, May 31, 2006 7:01 PM To: isp-clec@isp-clec.com Cc: isp-clec@isp-clec.com Subject: Re: [isp-clec] VoIP Feature Servers Brian, <flame> Sorry for the flame, but @ doesn't scale and has zippo in the support department. It is unbillable and unmaintainable. No less than four (4) other opensource PBXs have sprung from Asterisk because it is impossible to fix its problems: it is broken by design. These are carriers here, not hobbiests. Nobody wants to put their cashflow through uncertified, unwarrantied, piecemeal, castoffs. At least no more that 2 or 3 T1's worth. I could go on, but I'll shut up now. </flame> --brian On Wed, 31 May 2006, Brian C. Fertig wrote:I built one out of an open source platform called Asterisk. With it you can make it into what you want. 100% open source. Does TDM connections also.-- Brian F. G. Bidulock | The reasonable man adapts himself to the | bidulock@openss7.org | world; the unreasonable one persists in | http://www.openss7.org/ | trying to adapt the world to himself. | | Therefore all progress depends on the | | unreasonable man. -- George Bernard Shaw | To unsubscribe via postal mail, please contact us at: Jupitermedia Corp. Attn: Discussion List Management 475 Park Avenue South New York, NY 10016 Please include the email address which you have been contacted with. Copyright 2005 Jupitermedia Corporation All Rights Reserved.
Thread Index |
|
