x
VoIP Systems

Could SIP's Faults Foil Its Future?

Security problems and standards issues might wind up stalling SIP (Session Initiation Protocol) in the future, which would be a huge blow to IMS networks, according to the latest Light Reading Insider report, “The SIP Market: Trends, Opportunities & Threats.” (See Vendors Queue for SIP.)

For those of you just joining us, SIP is a text-based signaling protocol used to create sessions between endpoints in an IP network. It negotiates the type of session, the transport mechanism, and the encoding mechanism. In addition to other functions, SIP modifies and terminates sessions, and it's the primary signaling protocol used in IP Multimedia Subsystem (IMS) networks.

Since SIP lets users communicate in real time using more than one medium, users can make a voice call, then add a video session, then start sending instant messages to one another, for example.

But several things are threatening SIP's future.

One threat has to do with the singularity of the protocol itself, says Simon Sherrington, the Insider analyst who wrote the report. Standards like SIP need to be one thing to many people, not many things to many people.

Sherrington says vendors have a tendency to add functionality to their products in a way that embellishes SIP but runs ahead of the standards process. “This might gain a short-term competitive advantage, but risks greater interoperability problems in the future,” he writes.

Sherrington also points out that as developers build on the original SIP standard, the protocol becomes more complex. SIP’s initial appeal over other protocols was that it was simple and easy to implement. If some of that simplicity goes, so might some of the appeal, he suggests.

In a way, SIP could end up being a victim of its own success. Sherrington says more groups want to use SIP than were originally imagined. These groups -- like the wireless and cable communities -- are evolving their own flavors of the protocol.

“The requirement for SIP to work across diverse sets of networks could potentially create an unwieldy protocol for service providers that want services to be access network-agnostic." An even worse danger is that SIP’s variants become so different that they’re no longer compatible.

Security issues around SIP applications also continue to pose a threat to SIP’s large-scale adoption. While gravitating toward the utility of SIP applications, some carriers and enterprises have ultimately been scared off by the security risks. (See Covergence Banks on SIP Risks.)

As such, Sherrington's report says, security pros have been hard at work securing SIP and have made significant progress. For instance, Sherrington writes, signaling firewalls have been developed to protect against "security issues caused by overload, malicious attack, malformed messages, or irrelevant protocols."

Regulators also play a role in SIP’s fate. SIP applications, by their nature, depend on easy passage across networks and across borders. So an environment where carriers can exchange traffic simply and affordably is crucial. Without that, Sherrington says, the growth of SIP services could slow. (See NeuStar Moves Into SIP Peering.)

SIP was created by the Internet Engineering Task Force (IETF) in 1999. Sherrington says SIP applications will hit the market and drive revenues well before full-blown IMS applications begin to appear. Many carriers will develop and roll out SIP applications and services first, then use SIP as a foundation for a more complete IMS make-over later on. (See Ubiquity Launches SIP Program.)

Aside from regulatory, security, and standards threats, SIP does have several things going for it. Those are covered elsewhere in Sherrington's report -- find out more about it right here.

— Mark Sullivan, Reporter, Light Reading

jamesbond 12/5/2012 | 3:45:15 AM
re: Could SIP's Faults Foil Its Future? Well how do you explain MPLS/VPLS success then? In general I agree with you though.


materialgirl wrote:

This is why dumb networks rule. Complex interfaces always fail of their own accord. Keep the network simple, and put the complexity in the end-points.
materialgirl 12/5/2012 | 3:45:15 AM
re: Could SIP's Faults Foil Its Future? This is why dumb networks rule. Complex interfaces always fail of their own accord. Keep the network simple, and put the complexity in the end-points.
jayavenu 12/5/2012 | 3:45:11 AM
re: Could SIP's Faults Foil Its Future? The only way seems to be to ensure all major players get roped into 3GPP, TISPAN--ETSI,and PacketCable , so that they can agree to a universal SIP protocol and extensions.
Otherwise it will be a case of killing the golden goose for the IMS eggs!
paulej 12/5/2012 | 3:45:08 AM
re: Could SIP's Faults Foil Its Future? I simply cannot agree with the comment that dumb networks rule when it comes to services like voice. Even for services like e-mail, there is a lot of complexity. SMTP is relatively simple, but it still requires a servers in the IT department, quite likely with some external relay devices.

Too much intelligence in the endpoint means normal users (not the typical readers here) are going to be calling for support far too often. Interoperability problems will abound. There will be total confusion as users try to figure out how to get their phones to work and what phones to buy.

Voice services and other services needed by the masses need to be provided by application service providers. That might be Vonage or it might be AT&T. Whatever the case, permutations and combinations that lead to excess confusion equates to higher operational costs. That means poorer service.

Now, that does not mean I like IMS, either. SIP is a complicated animal and IMS turns it into a complicated monster.

Something else is needed that serves to stike the right balance. SIP was a good try, but not quite right.
standardsarefun 12/5/2012 | 3:45:00 AM
re: Could SIP's Faults Foil Its Future? jayavenu wrote:
"ensure all major players get roped into 3GPP, TISPAN--ETSI,and PacketCable , so that they can agree to a universal SIP protocol and extensions"

In other words: SIP is now too important to leave it in the hands of the theory boys and girls from IETF. Time to place it firmly in the smart network so we can all make some money!

P.S. remember everyone: dumb networks and smart end-user clients based on free download software means no money for anyone.
alchemy 12/5/2012 | 3:44:59 AM
re: Could SIP's Faults Foil Its Future? standardsarefun writes:
In other words: SIP is now too important to leave it in the hands of the theory boys and girls from IETF. Time to place it firmly in the smart network so we can all make some money!

P.S. remember everyone: dumb networks and smart end-user clients based on free download software means no money for anyone.


Right, so give it to the cellular network infrastructure companies so they can invent a standard so complex that:
a) a service provider is vendor locked
b) innovative & disruptive simplifying solutions are impossible

Just as Cisco maniuplates the IETF to stifle their competitors, the big telecom companies are doing the same at 3GPP.
HOME
Sign In
SEARCH
CLOSE
MORE
CLOSE