& cplSiteName &

Cisco Muddles on IMS

Light Reading
News Analysis
Light Reading
7/25/2005

Network infrastructure giant Cisco Systems Inc. (Nasdaq: CSCO) is taking a surprisingly evasive approach to the emerging IP Multimedia Subsystem (IMS) technology market, according to the latest Heavy Reading report.

Despite Cisco’s status as a leading provider of global IP infrastructure, the report -- “IMS and the Future of Network Convergence” -- finds that the company is yet to fully commit to a technology designed to deliver IP applications over different types of wired and wireless access networks (see IMS Guide).

“The company’s attitude to IMS appears equivocal,” writes report author Graham Finnie. “In a briefing with Heavy Reading in late 2004, the company was highly skeptical about IMS; the recent momentum behind it has led Cisco to fall more into line with the IMS approach.”

However, Finnie adds that recent media articles written by Cisco employees have been largely hostile to IMS, criticizing the technology as too complex, too costly, and with little benefit for end users.

Indeed, unlike rivals such as Ericsson AB (Nasdaq: ERICY) and Siemens AG (NYSE: SI; Frankfurt: SIE), which have taken a lead in the market by virtue of their early commitment to specific IMS products, Cisco has a range of pre-existing products that are being reworked for IMS.

“This product and others... are not strictly IMS products,” comments Finnie.

Finnie notes that Cisco describes its kit as “a family of SIP Proxies (including the CSCP) that are aligned architecturally with the IMS Core Session Control Functions (CSCF) and support specific CSCF interfaces and protocols.” Cisco claims that this family already supports most important IMS interfaces and that it will add additional IMS interfaces to these platforms over time, dependent upon customer demand.

Nonetheless, Finnie maintains that Cisco remains the lone incumbent yet to truly reveal its hand in the IMS space. “All are now strongly committed to IMS -- with the notable exception of Cisco.”

Opinions differ on the value of IMS and its likely impact on telecom. Have your say in the current Light Reading poll: IMS.

— Justin Springham, Senior Editor, Europe, Unstrung

(13)  | 
Comment  | 
Print  | 
Newest First  |  Oldest First  |  Threaded View        ADD A COMMENT
Page 1 / 2   >   >>
turing
turing
12/5/2012 | 3:07:20 AM
re: Cisco Muddles on IMS
I agree with djvbsl that IMS service creation from 3rd party software firms is an unlikely event. I think carriers really want it because it provides them a standardized (somewhat) architecture for telco-style SIP network with convergence to their wireless service (maybe), specifically with the ability to charge for service in a more secure/controlled fashion.

I think the cisco guys are internally at odds with that model - at least the dynamicsoft contingent (the technical camp). But the sales/marketing side wants to sell to carriers who shockingly want to earn money, thus the battle.
dljvjbsl
dljvjbsl
12/5/2012 | 3:07:20 AM
re: Cisco Muddles on IMS
from the IMS Crunch article:

By creating a so-called abstraction layer between applications and networks, IMS allows service providers to use any commercial software firm that can write to mainstream Java-based languages.


To be very polite about this, this is preposterous. Abstraction layers were a common part of any commercial CTI architecture. Java-based languages and APTs such as JAIN have been around for years. To think that these will ease the creation of services is to forget that their equivalents have been avaialble to commercial applicaion developers for at least 20 years.

Remember the old Microsoft saw that next year would be the break out year for TAPI. It is just that next year is always next year and not this year. TAPI allowed the developer to ring telephones anc collect DTMF digits. IMS will allow the developer to do similarly trivial things in teh VoIP world. Real applications of real customer worth demand much more.

IMS answers no real application problems. It purported answers are either old hat such as the appliction layer or vacuous arm waving such as Java-based languages. It does not contain an idea of what the real issues in large scale service composition are about.

dljvjbsl
dljvjbsl
12/5/2012 | 3:07:20 AM
re: Cisco Muddles on IMS
From the article:

However, Finnie adds that recent media articles written by Cisco employees have been largely hostile to IMS, criticizing the technology as too complex, too costly, and with little benefit for end users.

Cisco also states in the article that the conventional SIP elements that they are building fit into the IMS architecture. This ties in with my impression that IMS is a complex reworking of the basic SIP architecture that adds no customer value.
turing
turing
12/5/2012 | 3:07:19 AM
re: Cisco Muddles on IMS
With these caveats, IMS is a useful exercise. It cannot solve problems for which there is no current technological solution. Open service composition creates problems for which no one has solutions. It is better to do something real than to promise everthing and deliver nothing.

On the contrary, half of IMS' value is just putting to paper an architecture and requirements doc set, for how to build/design a voip service one can integrate into wireless and reasonably charge for. They threw in some more headers to handle some specific carrier issues not really addressed in previous SIP drafts (which is amazing considering how many pages of SIP RFCs/drafts there are)

As for the hype - I agree. The open services issue is a very hard one I think. From a technical and business aspect. Ignoring the technical, it's not really in the business interests of the vendors to let you buy service applications other than their own brand, so they have no vested interest to fixing their stuff to make it work with other vendors.

But sometimes you have to promise everything to implement the basics. And sometimes you get more than you thought you would. Take the case of SMS messaging for example - a service people (including me) thought would be a money-losing fad, but actually accounts for revenue. Or charging for ringtone and game downloads. Sounded like a horrible idea to me, but accounts for real revenue. So now they think if they can connect the cellphone game to your buddy's PC and let you play against each other for a fee, people will pay. (as an example)
dljvjbsl
dljvjbsl
12/5/2012 | 3:07:19 AM
re: Cisco Muddles on IMS
I think carriers really want it because it provides them a standardized (somewhat) architecture for telco-style SIP network with convergence to their wireless service (maybe), specifically with the ability to charge for service in a more secure/controlled fashion.


With these caveats, IMS is a useful exercise. It cannot solve problems for which there is no current technological solution. Open service composition creates problems for which no one has solutions. It is better to do something real than to promise everthing and deliver nothing.
turing
turing
12/5/2012 | 3:07:19 AM
re: Cisco Muddles on IMS
Cisco also states in the article that the conventional SIP elements that they are building fit into the IMS architecture. This ties in with my impression that IMS is a complex reworking of the basic SIP architecture that adds no customer value.

How does one tie into the other? Cisco's saying their products are virtually IMS compliant (which in marketing speak means they're not). Then you say that equats with IMS being a complex reworking of basic SIP architecture that adds no customer value? Just because cisco doesn't support it yet??

Of course I agree IMS is a complex reworking of SIP architecture - by definition whenever you take one architecture as a base and modify it you are "reworking it" and when you add stuff to make it do other things it will become more complex. And it may have little consumer value, but it might have real carrier/operator value - and they're cisco's customer.
alchemy
alchemy
12/5/2012 | 3:07:17 AM
re: Cisco Muddles on IMS
dljvjbsl extracts:
By creating a so-called abstraction layer between applications and networks, IMS allows service providers to use any commercial software firm that can write to mainstream Java-based languages.

My understanding is somewhat different from this. The core service proxy, the S-CSCF, accesses a database when it receives an INVITE and downloads a set of triggers to handle the call/connnection request for that subscriber. I believe the triggers are XML spec, not Java. This allows the S-CSCF to interact with various application servers and feature servers. It's like a soft version of what AIN was supposed to be.
alchemy
alchemy
12/5/2012 | 3:07:17 AM
re: Cisco Muddles on IMS
turing writes:
I think the cisco guys are internally at odds with that model - at least the dynamicsoft contingent (the technical camp). But the sales/marketing side wants to sell to carriers who shockingly want to earn money, thus the battle.

Give that man a cigar. You've captured the essense of the issue.

I ask the question:
Why would a vendor bother spending the engineering resources to build a DynamicSoft-style lightweight SIP Proxy core? It's not hardened (primary line, secure, five 9's up time) so service providers can't charge money for the services they offer on it. Therefore, they won't be willing to pay much money for those SIP proxies. That's why Cisco was able to buy DynamicSoft for chump change. It's a failed business model. A couple of bucks per subscriber just isn't worth entering the market.

The way to make money is for vendors to charge huge pile of money for complex gear inside the core. In the SIP world, that'd be session border controllers and the like. The DynamicSoft people view SBCs as the root of all evil. You'll notice that Cisco has no product offerings in the SBC space. In IMS-speak, the edge device called a P-CSCF is really a fancy SBC that translates, polices, and filters SIP signaling from the edge device. The Cisco/DynamicSoft dogma on this is going to prevent them from making much money on this part of the VoIP market.
dljvjbsl
dljvjbsl
12/5/2012 | 3:07:16 AM
re: Cisco Muddles on IMS

alchemy writes:

[in regard to IMS triggers or abstraction layer]

Its like a soft version of what AIN was upposed to be

The issue with this is that the AIN did not work and did not work for the same reason that the IMS will not work for the issue of applications. The AIN had no answer beyond arm waving for service composition. IMS does no better.
beowulf888
beowulf888
12/5/2012 | 3:07:15 AM
re: Cisco Muddles on IMS
Alchemy wrote:
"The way to make money is for vendors to charge huge pile of money for complex gear inside the core."

I got a chuckle from that cynical remark. The downside of IMS is that it looks to me (correct me if I'm wrong) to be a bloated vague marketing-driven standard that every carrier can meet -- except, of course for one crucial requirement of a true open standard: i.e. interoperability between different carriers' implementations (or should I say: "interoperability between carriers' different implmentations"?).

cheers,
--Beo
Page 1 / 2   >   >>
Featured Video
Upcoming Live Events
October 22, 2019, Los Angeles, CA
November 5, 2019, London, England
November 7, 2019, London, UK
November 14, 2019, Maritim Hotel, Berlin
December 3-5, 2019, Vienna, Austria
December 3, 2019, New York, New York
March 16-18, 2020, Embassy Suites, Denver, Colorado
May 18-20, 2020, Irving Convention Center, Dallas, TX
All Upcoming Live Events
Partner Perspectives - content from our sponsors
Multiband Microwave Provides High Capacity & High Reliability for 5G Transport
By Don Frey, Principal Analyst, Transport & Routing, Ovum
5G + Cloud + AI + Ecosystem, Opening New World of Video
By Samuel Chen, President, Cloud & Data Center Marketing, Huawei
All Partner Perspectives