Analysts and Consultants
on Provisioning July 2002 #1
to the first edition of the ViewsLetter on service provisioning automation.
This series of reports examines provisioning at all levels, from the chip
and component to the business management software. Your input is
THE APPROACHING PROBLEM
IN PROVISIONING DATA SERVICES
Imagine the public phone network still based on patch-cord switchboards
and human operators: the entire U.S. population working in central
offices couldn't begin to handle today's call volume.
Now look at data, whose traffic volume recently exceeded voice and
has a huge growth rate (compared to voice). The magic of compound
interest applies to data volume just as it does to your savings account,
which means that demand for data services at some point (relatively soon)
will exceed the capacity of current methods to provision those services
when customers want them.
A small example: Looking to add more Internet access bandwidth
into our office, we found a DSL company whose price and service seemed
fine, but the delay for installation was up to 8 weeks. Not only was
that a problem for us, but it certainly didn't help either the DSL company
or the Local Exchange Carrier to get their revenue flowing. The problem
extends to almost everyone, not just DSL firms.
And doesn't the delay say a lot about the cost to provision such a
service? Just count the truck rolls, the order entry minutes, the
network engineering man-hours, and equipment configuration steps.
Getting data service requires too much manual effort.
T-1 installations faced a similar situation in the 1980's. Originally,
a craftsperson dragged ABAM cable over distribution frames.
As T-1 became popular, delays in getting lines grew to exceed half a year.
Digital Access and Cross-connect Systems (DACS hardware) changed the install
from a cabling process to typing commands at a terminal--the number of
lines going in continued to rise but the installation delay came down to
a few weeks. Mechanizing a manual process (and competition, and fiber
capacity) also reduced the cost for a T-1 line by more than 80%.
Going into the future, the communications industry faces larger and
more complex issues than installing a physical-layer service like T-1.
Users want routing and switching as a minimum for commodity services like
Internet access. Carriers need to generate greater margins from functions
at higher layers of the protocol stack--up to applications (firewalls,
virtual private networks, virtual PBX, etc.).
A business served by a local loop of copper or fiber can define a range
of data services including the bit rate, transmission formats, and higher
layer services--automatically--and immediately gain the use of those services.
The local loops (fibers or copper) are not defined as to capacity or service.
The customer premises equipment is standardized, so the brand used by
the subscriber need not be the same as that installed by the carrier.
While not a perfect comparison, consider a renter who moves into an
apartment and uses existing wiring for a phone--the customer can bring any
brand of phone and may define a wide range of services including caller
ID, distinctive ringing, voice mail, call waiting, conference calling, etc.
The setup of these services is not yet fully automated for the residential
market, but web-based customer services are moving toward that goal.
COVERAGE OF THIS VIEWSLETTER
We will look at all aspects of the provisioning problem, reviewing
and reporting on hardware and software products, standards, and regulations
that relate to the Vision. We welcome contributions and suggestions
in the areas of:
-- Chip technology such as network processors and all-feature interfaces.
-- Network elements with flexible functions on the user ports.
-- Management software that facilitates automated provisioning,
= element managers
= mediation platforms
= workflow integration systems
= carrier operations and support systems.
-- Government regulations and mandates.
-- Customer needs that justify the enhancements to network infrastructure
needed to support automatic or subscriber-initiated provisioning.
-- Economics of carrier provisioning.
We recognize the value of your time, and will aim for a length of two
to three pages per issue, at a frequency of about twice a month. Often
our web site will offer additional detail or related information, such
as illustrations and figures.
-- Visit our web site
for information about our consulting and analyst services, speaking engagements,
and updates on other activities, or contact us by email.
-- Special thanks to www.webtorials.com
for hosting our web site and ViewsLetters.
"Flanagan Consulting" and "ViewsLetter" are
Service Marks of W. A. Flanagan, Inc.
Updated: 11 June 2003