T1, Anyone? Data Centric CLECs Make the Push for Non Voice Services [ASND reference]
x-changemag.com
By Gary Kim
Only three things matter to competitive local exchange carriers (CLECs)
choosing access platforms: data, data and data.
"Voice over data is a huge trend," says Lonnie Martin, ADC
Telecommunications Inc. senior vice president.
Already, "many CLECs are focusing heavily on data services, including
ICG Communications [Inc.], e.spire [Communications Inc.] and MDC
[Communications] in Canada," says Roger Koenig, CEO of vendor
Carrier Access Corp. And that means growing emphasis on Internet
protocol (IP), asynchronous transfer mode (ATM) and frame relay
services.
But the practical realities are that CLECs are scrambling to provide the
services smaller businesses know they need today. That means support
for office key systems, private branch exchange (PBX) equipment, local
area networks (LANs) and Internet access.
So CLECs must connect to routers and voice switches. And, generally
speaking, that means a heavy reliance on T1 technology as the access
method.
Customer parsimony and carrier desire to maximize revenue are factors
leading to the reliance on T1. Since smaller businesses are quite price-
conscious, CLECs want low-cost solutions and the ability to capture
multiple revenue streams. That, in turn, leads to a desire for voice and
data delivery over a single pipe.
"I see the majority of CLECs going to a multiline data and voice platform
to address the smaller and medium businesses, those that have
$1,000-a-month total telecom bills, on a profitable basis," says Koenig.
The logic is fairly simple. CLECs excel financially when a single T1 pipe
can be filled up with voice and incremental data traffic. "One
infrastructure, multiple revenue streams" is the mantra.
Leasing full T1 circuits from the incumbent local exchange carrier (ILEC)
is an option. But CLECs can do even better when two copper pairs can
be leased at $50 and outfitted with high-bit-rate digital subscriber line
(HDSL) modems to deliver multiple voice circuits and high-speed
Internet or data access. So CLECs are installing traditional time division
multiplex (TDM) access equipment, including channel service units/data
service units (CSUs/DSUs) and integrated access devices (IADs) that
handle IP and frame relay traffic, typically using HDSL. Despite the
growing importance of newer synchronous protocols, T1 and DS-3 are
the mainstays of CLEC access.
"You can put lots of things inside that kind of pipe, which is why it's
popular and will grow for long time," says Martin.
Of course, there's no disagreement about the importance of IP and
ATM. But "the speed at which the data revolution occurs is the subject
of debate," says Stephen Susina, marketing manager for Tellabs
Operations Inc. "But the traditional transport market continues to grow,
and there still is no shift in customer demand in terms of what they
actually deploy."
Packet Fever
Early summer telecom developments reinforced the notion that data is
hot.
Even media companies further validated the notion that the Internet is a
strategic new medium when Walt Disney Co. bought a 43 percent stake
in Infoseek Corp., the web browser/
portal firm, with an option to take majority control. Meanwhile, the
National Broadcasting Co. (NBC) acquired 19 percent of Cnet Inc.,
another portal firm, with an option to increase the stake to 60 percent
within three years.
Telephony's equipment giants similarly signaled data's exploding
importance. Northern Telecom Ltd. acquired Bay Networks Inc. Lucent
Technologies rolled out its line of "PacketStar" IP systems, while virtually
every other major telecom switch provider said it would do the same.
Carriers also made data-centric moves. Sprint Communications Co.
announced an accelerated move to an all-cell, all-distance ATM network
it calls the Integrated On-Demand Network (ION). AT&T Corp.,
meanwhile, introduced its new "GeoPlex" software environment, which
allows third-party programmers to develop applications for the AT&T
network. AT&T also tried to buy America Online Inc., but was rebuffed.
It's not hard to figure out why data is the "only" thing that matters.
According to analysts at Strategic Networks Consulting Inc., IP traffic
will consume more than 90 percent of global bandwidth by 2003.
Planners at Bell Atlantic Corp. and Sprint likewise estimate that IP
revenues will eclipse 1998 voice and data revenues by 2001 or so. And
of the 80 billion minutes worth of U.S. network traffic in 2001, about half
will consist of IP or frame relay packets or ATM cells, according to
Yankee Group analysts.
In the next century, data services will provide 80 percent of carrier
revenues, but will comprise only 40 percent of network traffic, says
Voorhees, N.J.-based consulting firm CIMI Corp.
That has to drive CLEC thinking on access platforms as well. Yankee
Group researchers are more cautious, estimating that of the $140-plus
billion carrier services market in 2001, just shy of $40 billion will be
contributed by data services. So it stands to reason CLEC access
thinking will be dominated by data access.
T1
Of course, the U.S. competitive access provider (CAP) market was built
on private line (T1, 1.544 megabits per second [mbps] connections,
especially) service, primarily to connect business customers with their
long distance carrier points of presence. But CLECs now are assaulting
the much-larger switched services market. And since data increasingly
drives the business, frame relay, IP and ATM services should be wiping
out the private line T1 business, right?
Nope. T1 continues to grow. Today, some 1.6 million T1 lines are in
service in the United States, with growth rates in the 25 percent-a-year
range, according to researchers at DataQuest, a Gartner Group Inc.
company, and the Multimedia Telecommunications Association. This
figure includes both "wholesale" use of DSL as a method for increasing
capacity on existing copper and "retail" lines sold to end users.
Executives at PairGain Technologies Inc., for example, suggest that
DSL-driven T1 growth is in the 40 percent annual range.
Internet service providers (ISPs) are part of the reason T1 growth hasn't
leveled off the way many observers anticipated.
"To support the growth of the Internet, 350,000 T1s and 25,000 T3s will
need to be provisioned during the next four years," says Paul Johnson, an
analyst with banking concern Robertson Stephens. "Compare this with
today's installed base of about 300,000 T1s and 2,200 T3s."
DSL is another reason T1 keeps growing. Where traditional T1 systems
require repeaters every 3,000 to 6,000 feet, DSL loops can reach
12,000 feet without using any repeaters. That translates into lower cost
and higher usage. CLECs, for example, can lease "dry" copper pairs and
convert them into T1 spans using HDSL modems.
"The most important thing today is to aggregate voice and data over a
single pipe," says Dave Gallerman, Newbridge Networks Inc. vice
president. "And of course, price, price, price."
The emphasis on integrated, affordable solutions is a direct result of
CLEC focus on smaller and medium-sized businesses. Such customers
focus on upfront capital costs, not life-cycle cost or "total cost of
ownership." They also tend not to have in-house networking expertise.
So CLECs want a simple, do-everything solution and equipment prices
that don't scare customers away.
None of which is to say demand for capacity won't skyrocket further.
"DS-3 prices are going through the floor now," says Jesse Price, Eastern
Research Inc. marketing vice president.
If so, CLECs will be looking for channelized DS-3 equipment that
terminates directly at the PBX and router, says Price.
As that trend continues, CLECs will gain more freedom to choose
platforms. Today, it makes a difference which transport format a carrier
chooses.
"Up to about 9mbps, it makes a cost difference whether a carrier
chooses ATM, frame relay or TDM," says Tom Nolle, president of
CIMI. "At T3 and above, it doesn't matter that much."
What ultimately may matter is which packet format is used and by whom.
Most incumbent carriers are ATM adherents. ISPs and several of the
new long distance carriers are IP-centric. It remains unclear what CLECs
ought to do.
"The crux of the CLEC business is that it's hard to make bets about
technology," says Leif Hoglund, business development vice president for
E/O Networks.
T1 is fundamental. "But the access part of the market is really fraught
with confusion," Hoglund says. "ATM to the desktop, IP over SONET
(syncronous optical network) are examples. You can't tell where it's
going to go."
What is abundantly clear is that no carrier wants to deal with "stranded
capital," in which network assets are deployed but only a fraction of full
capacity is used, or in which capacity cannot be expanded to meet
additional need. Even the regional Bell operating companies (RBOCs),
for example, now are investing in smaller line-count digital loop carrier
(DLC) equipment, says Kris Sowolla, E/O Networks product
management director.
CLEC Trends
CLECs tend to fall into three camps when it comes to access platforms.
Some carriers have a clear focus on services they expect to offer, and
want an optimized platform that addresses those services. Such a
business strategy favors single-purpose terminals that trade flexibility for
lower initial capital investment. Chicago-based Focal Communications
Corp., for example, is a local dial tone specialist. NorthPoint
Communications Inc. and Covad Communications Co., have different
business strategies based on a wide variety of potential services, including
voice, data and local and wide area networking. In such environments,
flexibility is more important.
Some contestants also are betting that new applications, as yet
unforeseen, will be important. So platform adaptability is strategic for that
reason, as well. Data-centric and "bundled services" providers such as
Intermedia Communications Inc., GST Telecommunications Inc. and
Convergent Communications Inc. are examples.
"If you're in a 'get-me-to-revenue' mode, you may not want to pay for a
multifunction solution," says Gallerman.
The low-cost, application-specific terminal fits there. But "some service
providers want to do a limited set of things real well, while others want to
do everything the customer wants," says Gallerman.
The integrated, feature-rich device may work better in that scenario.
Its the Equipment Stupid
As one might expect, two concurrent trends are seen in access
equipment. Cheaper, single-purpose equipment and integrated,
multifunction devices are available. One example of the single-purpose
terminal is the ADC Satellite 651 CSU/DSU. It's aimed at the
business-to-business ISP, for example.
Vina Technologies Inc.'s T1 Integrator is a prime example of the
multipurpose trend. The Business Office Exchange version, for example,
supports key system voice switching as well as remote and Internet
access. It incorporates the functions of a router, channel bank, T1
CSU/DSU, multiplexer, frame relay access device and Internet firewall.
Adtran Inc.'s Total Access Multi-Service Optical Platform is another
example of the "multifunction" trend. Total Access supports HDSL,
integrated services digital network (ISDN), T1, optical DS-2 and DS-3
access at a central office or point-of-presence. Likewise, Premisys
Communications Inc.'s Q-155 supports TDM traffic (T1/DS-3), frame
relay, IP, ATM and HDSL on the access side of the terminal, and
SONET on the transport side of the terminal.
HDSL integration also is a prime trend. Telco Systems Inc.'s Access 60
and Access 45 integrated access devices, as well as the ADC
Telecommunications/Carrier Access Corp. EZT-1/DI-HDSL terminal,
for example, incorporate HDSL directly into access terminals. One
advantage is the marrying of Soneplex management tools with the
lower-cost access device.
"We're seeing more and more requests for simple network management
protocol (SNMP) capabilities built right into the access vehicles," says
Rick Vesny, ADC marketing director.
Carriers also are trying to collapse whole layers of equipment in the
network. Trends such as IP over SONET (see related story, page 26)
and multiservice switches are examples. The Ascend Communications
Inc. GX 550 combines digital cross-connect system (DCS) with SONET
multiplexer functions to collapse a SONET, DCS and frame relay/ATM
switch into one device.
The same approach to simplifying networks applies to the Lucent
Technologies Inc./Sun Microsystems Inc. "ISP-in-a-box" approach. The
two firms bundle IP telephony, remote access, network management,
e-mail server, web server and electronic commerce capabilities.
IP was once considered a legacy technology destined for rapid
replacement by open systems interconnection (OSI) protocols. Instead,
OSI died, while IP has swept the world. T1, also a legacy technology, is
probably going to surprise us as well.
Gary Kim is strategic research director for Convergent
Communications Inc. He can be reached at (303) 749-3061.
Copyright c 1998 by Virgo Publishing, Inc.
Please read our legal page before using this site.
|
