More SIP Trunking Gotchas
Number porting, Caller ID, E911, "bursting," and other issues can complicate and slow down your migration.
Man, this SIP Trunking issue just keeps rolling along. Last week, we held a "virtual event" on SIP Trunking--6 webinars back to back, plus vendor spaces where attendees could gather information from the virtual event's sponsors--Genband, XO Communications, Cisco, Oracle, Sonus, and Level 3. Pre-registration was among the highest we've had for this kind of event, and attendance likewise was substantial--almost 400 people listened to our keynote webinar from Jim Allen and Dave Stein, the independent consultants who serve as our SIP Trunking gurus.
As we've seen in the past, the topics that drew the most attention from speakers and questions from the audience centered around the real nuts-and-bolts issues, the "gotchas" that can trip up a SIP Trunking implementation. Some of the challenges that XO highlighted in one webcast included:
* Porting: To ensure that numbers are ported to the new access trunks without disruptions, enterprises need to make sure they have an accurate inventory of current DIDs, and that if they're switching carriers, they're prepared to deal with, shall we say, less than enthusiastic cooperation from the carrier losing the business. The bottom line: Start addressing this issue early in the process and understand the likely pitfalls.
* Caller ID: Because SIP Trunking allows multiple locations to sit behind a single, centralized SIP Trunk, you may encounter problems passing accurate Caller ID information through to the called party. Our XO panelists pointed out that, in the PSTN world, the physical location of the caller--the originating LATA--is always known, and this information can always be used by the carrier to deliver accurate Caller ID. In contrast, the SIP Trunking carrier may not know the physical location of the caller, so the system must be configured to make sure it can pass the Caller ID for the DID in a way that can be read at the remote end. (A variation on this is the challenge of E911 and making sure that a call reaches the Public Service Answering Point or PSAP that serves the physical location of the caller.)
* Fax, modems, and alarms: These legacy systems will remain in place, so enterprises have to plan carefully to make sure they can make the transition. Fax is an ongoing issue that more people are talking about, but our XO speakers noted that if analog modems can't be phased out, you may need to retain POTS lines to serve them--i.e., there is no solution within SIP Trunking at this time.
* "Bursting": By now, many carriers offer a service in which enterprises can flexibly use the capacity they're buying--in other words, instead of being restricted to a given amount of bandwidth at each location, with the risk of overprovisioning at some places or times and underprovisioning at others; the carrier may offer a "pool" that lets you reassign capacity dynamically so that you optimize your bandwidth utilization. But, as Steve Lingo of XO has warned in the past, these "bursting" offers vary among carriers, so you have to understand what you're buying.
We keep waiting for SIP Trunking to become commoditized and simple to adopt and implement--at which point, it will presumably no longer be a hot topic. That keeps not happening, because there seems to be an endless layer of challenges and problems that need to be solved before you can implement SIP Trunks and capture the savings they promise, while not having them alter your telephony environment in ways that hurts the business.
The bottom line is that a SIP Trunking-for-PRI swapout is not as straightforward as an MPLS-for-frame-relay swapout. Getting the benefits without losing the features is a much more detailed, tricky process. As long as that remains the case, SIP Trunking will be a big draw.
By the way, if you missed the virtual event, you can view all the webcasts and see the sponsor materials on demand.