|
Archive
Flanagan Consulting
Network Analysts and Consultants
"We Have the Experience"
---------------------------------------------------------------------------
ViewsLetter on Provisioning
24 March 2003
#19
---------------------------------------------------------------------------
>>Analysis
NET.COM WANTS IP SERVICES TO BE MORE LIKE VOICE:
ON-DEMAND, WITH GUARANTEED QoS/SLA, AND PROFITABLE
--By William A. Flanagan, Editor and Publisher, ViewsLetter
"You can't build an infinite capacity core for the Internet," says Bert
Whyte, CEO of net.com ( www.net.com ). "Unlimited bandwidth doesn't
work--you need to charge for usage. There's an obvious guide in
the
public switched telephone network."
In a recent interview, Whyte pointed out that voice still represents more
than 80% of the revenue in telecom, and enjoys high margins. IP
networks,
though they carry more than half the traffic, suck up dollars rather than
throw off profits. What's the problem?
"IP people don't understand the business model of phone companies, which
is a bandwidth business," claims Whyte. "They don't respect carriers
and
have no appreciation for the 'voice' culture and the long history behind
telephone service. They offer a Local Area Network model, but it
won't
work for an IP Wide Area Network."
IP carriers don't know what's on their LAN-like networks, so a few big
users can monopolize the network capacity, even for illegal uses, and
pay
no more than the casual residential customer. The result is an "obscene
model" for the business.
To get IP profitable, Whyte believes the public voice network sets a good
example:
-- service on demand (dial your own calls).
-- guaranteed quality of service (traditionally from dedicating network
resources exclusively to each call by each customer, QoS on IP nets).
-- accurate billing for the usage (and time, bandwidth, QoS, etc.).
"Broadband data services need to go the same way," says Whyte. "And
it
will be the edge device, the access platform at the carrier edge, that
sets each customer's QoS and features."
The edge device needs to protect the network from users who can't pay
for
what they consume and from traffic that harms the network. But today's
IP
networks can't drive policies and QoS to the users, he says. To
make this
solution practical, the process must be automated.
To put more effort behind automating data services, Bert helped found
the
Service Creation Community (www.servicecreation.org). Its members'
goal
is to make carriers more profitable (OK--and to sell some of their own
stuff in the process). The group includes hardware vendors, Operations
and Support Systems (OSS) software companies, content providers, and
carriers. They're focused on defining reference models for
revenue-generating services, which thus far include video on demand and
user-selected class of service, using members' expertise, equipment, and
software.
"The SCC was needed to balance the long-standing tendency of carriers
to
build single-vendor networks," said Whyte. "Because carriers did
it for
voice, they've easily adopted one vendor for IP networks. And while
that
vendor supports standards, it also always adds an extra 'extension' or
otherwise puts its own 'spin' on an implementation. That makes
interfacing to another vendor's device tortuous or sometimes impossible."
So the SCC's work is to open the carriers to multi-vendor, best-of-breed
solutions that let in more creativity in less time. The legacy
single-vendor voice switch, for example, evolved very slowly--for several
reasons. The vendors certainly wanted to ensure each released product
was
stable and ready for the PSTN. They also had time--the closed
architecture prevented anyone else from supplying a feature or function
that a carrier wanted.
That's not the way forward to profitable IP networks. To get the
best
products and the lowest cost in the shortest time, public networks (voice
as well as data) need to be open to whoever has the best solution.
Carriers need the wide-based innovation to create products (ASAP) that
generate new revenue with good margins. That means applying standards
fairly strictly, encouraging vendors to make accommodations for mutual
interoperability, and avoiding the FUD Factor (Fear, Uncertainty, and
Doubt used as a selling tool).
IBM invented FUD, for mainframe computers. They've left it behind
(or
passed it to another vendor) and now offer added-value in standards-based
servers. ATM went through similar phases, for a time doing compatibility
testing at the "mid-span meet." IP-based products need a similar
process
to bring a maturity and stability to packet networks that will assure
carriers they will have the resources and flexibility to bet their futures
(individuals' jobs as well as companies) on the next generation
technology. A toast to SCC.
============ + ===== + ===== + ===== + ============
MORE LINKS
-- Nicole Drapeau Gillen ( ngillen@greenwichtech.com ) contributed the
"Dashboard Dancing" piece in the last issue, where we failed to include
the link ( www.greenwichtech.com ) to Greenwich Technology Partners, where
she is a Principle. Thanks again to Nicole for the article.
-- Visit www.flanagan-consulting.com when you need independent review
and verification of network architecture, product positioning,
or
your marketing message.
"Flanagan Consulting" and "ViewsLetter" are
Service Marks of W. A. Flanagan, Inc.
Updated: 11 June 2003
|
