Bandwidth: Issue for All

"Rather

than business driving bandwidth, in India, we have bandwidth

driving business," said a participant in a panel discussion

organized by Voice & Data recently. And he was

more than right. We found that out when we spoke to selected

users for finding out what were the issues that daunt them the

most. Bandwidth not only figured prominently as an issue, quite

a few other issues also emanated from what is primarily a lack

of bandwidth problem.

For example, many users

feel the reliability of new applications is an issue that delays

decisions. But most applications that they mention are quite

mature globally and are deployed by large number of

organizations. On closer scrutiny it is revealed that it is not

really the application "reliability" that bothers

them, but the "speed" of the application that they are

worried about. The culprit: Neither the application nor the

integrators. It is bandwidth. This is too common. Many companies

today delay application deployment because of bandwidth

constraint. Those who have been more optimistic and have gone

for the new bandwidth-hungry applications on wide area have

suffered.

Today, many organizations

have implemented high-speed LANs. A large number of them have

gone for Fast Ethernet and quite a few for Gigabit. In the

applications that are restricted to local area, this has eased

down things. But it has created more problems for the wide area.

Users now want to use these applications on a wide area basis.

Internet access is now available to all. Many organizations have

the facilities today for Internet access through the



LAN by almost everybody in the organization. A 128 Kbps
leased-line is used sometimes by as many as 150 users.



You can imagine the speed! A simple e-mail application gets
stuck.

But we Indians are very

familiar with scarcity. What really hampers the market growth is

the user's perception. As the oft-repeated saying goes, the

maximum bandwidth that you have is the bandwidth of the thinnest

pipe, anywhere in your network. In India, that thin pipe is in

the middle. We have global bandwidth providers like SEA-ME WE2

and FLAG, who have huge capacities available with them. Project

Oxygen is also focusing a lot on India.

That is on the global

side. On the enterprise side, the decision is really in the

hands of the users. It is the national bandwidth availability

that is an issue. Though the DoT says it has enough bandwidth

available with it, few believe that it really understands the

magnitude of the requirement. The perception stems from the fact

that getting bandwidth from DoT is a project by itself. Getting

a 2MB leased-line today takes so much of effort. As a result,

many users today refuse to take the risk of technology and

applications as they feel bandwidth will not be available in

time for them to run their applications. Despite the bullish

outlook projected by researchers for the telecom market, many

users refuse to believe that situation will improve drastically

even in the next few years, let alone months.

More grave is the

situation of new investors who are looking at India for the

investment of remote back office and other communication-enabled

services like call centres and Internet data centres. Says P

Swaroop, head of IT at Hewlett-Packard India on record, "HP

wants to set up a global contact centre in India. But we find

there is a severe bandwidth issue." In the long run, this

will not only hamper India's dream of becoming a global IT

superpower, but will also hamper the employment generation from

these services. The NASSCOM-McKinsey study, for example,

estimates that there might be an employment loss of about

3,50,000 in IT-enabled services alone, because of the lack of

bandwidth.

The Big Picture

Everybody's

problem is nobody's problem. Even today, all the industry

lobbying is centred on duty structures, licence conditions, and

terms of agreement. If bandwidth is mentioned at all, it is not

in any significant way.

A $50 billion Indian IT

dream is what even school children today talk about. But how

much of that is achievable with this kind of bandwidth? The

NASSCOM-McKinsey study estimated an export revenue loss of $22.5

billion because of the lack of the availability of bandwidth.

This includes the loss in software exports, IT-enabled services,

e-businesses, and IT services. "Moreover, in the domestic

market, Indian users will be unable to deploy state-of-the-art,

global networking applications/products that will be developed

for high-speed telecom networks. India will also be constrained

in utilizing its supply base of knowledge workers, especially

those operating from homes (who would require telecom to receive

inputs and deliver outputs)," the study further finds out.

And remember this study

was undertaken at a time when the dotcom revolution in India was

yet to begin. With VC money in abundance, the dotcom companies

have bigger plans today than the major IT and telecom companies.

And that itself is a mismatch. Applications are going to be

built much faster than the bandwidth. Add to that the

multimedia-communication/entertainment revolution factor. There

you start and end with bandwidth.

One was hoping that ISPs

would at least bring in some focus in terms of better

utilization of bandwidth, if not provisioning of bandwidth. But

barring a few, many went after numbers. Says Balakrishnan R,

COO, Euclid India, "Soon ISPs will have to play a critical

role. Customers are looking at a purely outsourcing kind of

scenario, when they will move from 'how much of bandwidth'

to 'what is the performance'. They will ask for bandwidth

guarantee." At least for Internet access, which is a major

bandwidth consumer, the new private gateways is likely to solve

some problems.

But it is also the local

access that matters. DoT leased-lines are not reliable and the

service is not at all professional. Going for high-bandwidth

applications is more of an adventure. Digital Subscribers Line (DSL)

and cable have been fighting a technology battle in the

broadband access in India. Whichever wins, the victory is the

user's. Most private telcos, including Tata Teleservices in

Andhra Pradesh and Hughes Ispat in Maharashtra have gone for DSL-based

access provision. Bharti Telenet in Madhya Pradesh and Telelink

in Rajasthan are also testing this technology. Summary: Indian

p-telcos are betting on leading-edge technologies to solve the

problem as much as they can.

But what is really a

milestone is the step that two companies–Spectranet and

Hathway–have taken. The companies have been building local

high-speed networks in Mumbai and the National Capital Region

(NCR) respectively. These companies are targeting the corporates

as well as households. They may be able to make a significant

difference in one area:



access.

The other significant

development is long distance networks. There are quite a few

players who are serious about the long distance market in India.

Most of them are planning high-speed Dense Wave Division

Multiplexing (DWDM) networks. Though no one knows when

things will get rolling–the DLD policy is not yet clear–things

certainly look brighter.

India, however, has a long

way to go before its IS managers stop worrying about bandwidth.

And that is just one part of the story, which is the scope of

this issue.

The real bandwidth problem of India–if

you include the common subscribers and planned government

networks–is much more severe that this.