TeraBurst steps into optical jungle
The company announces its first products just as times have become lean and mean for optical equipment makers.
The Sunnyvale, Calif.-based company introduced two optical switches, the OMS 2200 and OMS 2100, that manage and transmit communication signals for telecom carriers. The 2200 is meant to help transmit data over long-haul networks, or networks that run between cities, while the 2100 is a smaller box targeted at networks in urban areas.
These products are being offered during a time when telecom carriers are planning to buy less equipment in general and, according to some analysts, even less from newer companies. TeraBurst is privately held and only started in January 2000. The start-up managed to haul in $51.4 million in funding from investors like Crest Communications Holdings, Labrador Ventures and Merrill Lynch Ventures.
"It's clearly tough for any company at this point, particularly start-ups, because future investments are being made with more scrutiny," said Tim Smith, an analyst at Gartner Dataquest.
Other analysts agree, and point out that telecom carriers want to go with equipment they know in tough times rather than take a more experimental approach and try out newer equipment.
TeraBurst's switches run on an O-E-O (optical-electrical-optical) design, which means that signals enter the switch in light form, become converted to electrical impulses that are then reshaped, and then get converted back to optical signals and sent out.
The signals need to go through this process because sometimes the data needs to be broken down and sent in smaller packets to different destinations. Shin Umeda, an analyst at Dell'Oro Group, says the process is comparable to the role of a retail wholesaler. The wholesaler receives goods in mass orders from the factory then takes the task of dividing the shipment and sending it out in different volumes to retail chains or mom-and-pop stores.
O-E-O switches are slower than all-optical switches because of the electrical interlude, but TeraBurst has expedited the electrical phase by handling that part of the process differently, and claims speeds that approach all-optical switches.
In fact, the start-up does not think that all-optical is the way to go.
"There (have) been a lot of marketing dollars spent on positioning (all-optical) as the Holy Grail, but our position is that is not necessarily true," said Shantanu Mitra, director of product marketing at TeraBurst.
Shantanu argues that a signal is much easier to work with in electrical form because it can be strengthened and monitored while the technology to do this to light signals has not arrived yet.
Still, Umeda and many other analysts say that other companies like Ciena, Nortel Networks and Sycamore Networks are attacking this problem from other angles and it will be awhile before telecom carriers crown the winning technology.
Analysts now say that TeraBurst has to execute its business plan, which aside from making sure the products work, involves manufacturing it in volume and persuading telecom carriers to use it.
The company does claim some promising developments on this end, and says that at least two telecom carriers have agreed to test it, which is the first step to possible purchases down the road.
Michael Howard, a principle analyst at Infonetics, thinks the technology has promise, but also says that a small company like TeraBurst may have other options if it cannot compete against the established equipment giants.
He notes that Ciena, Nortel and Cisco Systems all carved a niche for themselves in the optical market by acquiring companies with the right technology, which might be a kind of happy ending for TeraBurst.
"Today, the reality is you either IPO or get bought, or IPO and get bought," he said.