 | |
 |
Weekly Start-up: Native Networks Technologies From Israel's Business Arena - Globes
July 31, 2000
Several months later, again over a drink, this time accompanied by a plate
of pasta, Goren decided that the time had come for him to leave the warm and
cozy concern and set up on his own. He invited over his friend Menahem
Kaplan, with whom he had once worked at ECI.
What did the pair decide to do? Solve the bottleneck problem at the access
point of fiber optic networks, i.e. the stretch between the exchange and the
customer. This is the point in which service providers encounter problems
vis-a-vis less than medium-size customers: traffic to these customers does
not warrant laying expensive and dedicated communications lines for a
sizable bandwidth. On the other hand, these customers' requirements change
frequently.
The less-than-medium-size customer may have scarcer resources, but his
business needs are just as important. He wants to balance between his
requirements and his most expensive resource, named bandwidth, allocated to
him by his service provider. He wants to optimize use of this expensive
resource to avoid a major increase in his communications costs. The service
provider, is of course anxious to please, without losing competitiveness and
without having to lay additional expensive infrastructure. In other words,
the service provider has in interest in maximizing the given infrastructure
in order to provide increasing customers with better service.
One of the methods for maximizing traffic is known as SDH (SONET in North
America), real-time multiplexing. According to Goren, this method does not
exploit the network resource as it should. In effect, it provides a pipeline
of a fixed width, insensitive to an organization's changing needs.
Native Networks Technologies decided to take the method of statistical
multiplexing which we have known for three years now, and apply it to
access. The company then added two innovations: to "transport" the network
protocols naturally, not by ATM which requires everything to be translated,
which shoots up maintenance costs. The Israeli start-up does it naturally.
The second innovation is at the level of bandwidth management. The
assumption is that the customer "rents" bandwidth from the provider and pays
for it. He therefore has undisputed priority on the resource. If the
customer does not have a certain type of traffic, first of all the bandwidth
is routed for his other applications lower down on his priorities list, and
only then is the resource relocated to other customers.
Despite Goren's training in theoretical physics, the company's technology
includes no new physics. "It's all a matter of algorithms, he says. The
product is scheduled to be completed in the first quarter of 2001, and then
tested in field trials in the second quarter."
Native Networks has not reinvented the wheel. The start-up has some
competitors: Appian, which has just announced its product, and Alidian,
which developed a similar product a different stretch of the network,
similar to Chromatis, and is currently developing a further solution. Goren
admits that Appian is a true rival.
"Globes": Aren't you late in the market?
At the seed stage, Native Networks raised $1.3 million from Israel Seed
Partners, SkyPoint Capital of Canada and Apax, Israel, at a company value of
$4.25 million, after money. The company is currently completing a $10-16
million financing round, at a company value that it is not yet willing to
disclose.
Published by Israel's Business Arena on 31 July, 2000
|
