Although there is a lot of significance in certain quarters being attached to AT&T’s trial of Coriant’s 400G system (as we recently pointed out on on fibeReality’s LinkedIn updates page), the engineers formerly with Tellabs and Nokia Siemens Networks (NSN), which make up the heart of the private equity-financed vendor, do not seem to have a great deal of familiarity with being a leading player at higher data rates for traditional long-haul and short-haul service providers. Tellabs itself was a metro transport player with its primary focus on lower speeds including optimizing around 10G as well as delivering very efficient 10-gig services, while allowing for the co-existence of 100G services within the same channel plan. While NSN made some penetration with 100G technology, and engaged in activities with some of the European telecom providers involving hero experiments at 400G, there was no significant differentiation with any of its optical product line, as it evidently shopped its optical assets to every vendor, and obviously, there were no takers. Moreover, the idea that any publicized trial of new equipment is being promoted so prominently (especially relating to AT&T) by certain analysts and by some members of the press obscures the headway being made by a couple of other suppliers, which are not being as vocal about their accomplishments.
There is an additional amount of irony in that Coriant had been a big supplier of 40G to AT&T, a carrier which had its line-side network operating at that data rate at an order of magnitude higher than at 100G in the first quarter of this year. In fact, our impression is that in recent years, Coriant had been extremely realistic in its low expectations for both market growth and technological developments beyond 100G in public networks, particularly for serial 400G, with the infrastructure hitting the tradeoff between reach and capacity. Clearly, while its data center interconnect platform has a total capacity of 400G, just as other vendors in that sector offering pizza boxes with relatively limited capabilities, the applications have been for far less amounts of bandwidth allocations.
At a time when hardware differentiation on transport systems is becoming increasingly more difficult, Coriant striving to stretch itself beyond the confines of its cultural foundation, even if it turns out that it is only providing a 2x200G for the trial, is an indication that Marlin Equity Partners will not be patient forever. The investment firm clearly misread the market with its acquisition strategy that created the present-day vendor, and is evidently applying lots of pressure on the company to separate itself from the pack.
To whatever limited extent the 400G space develops in the foreseeable future, Ciena certainly will be ready. We have a high level of confidence that Huawei has a serial 400G in the lab. According to our recent report, Clash of Optical Component Vendors & Technologies in Data Center Networks, Nokia Alcatel-Lucent has even done first-office applications on its single-lambda 400G device.
The other aspect that is amusing is the level of importance given to trials of new gear by both consultants and news people, which can often be disasters because there needs to be an opportunity to iron out all of the bugs. Even putting aside the ridiculous notion that AT&T will turn out to be a leading force on 400G, the immediate reaction to Coriant’s equipment being tested should automatically make it a potential supplier of such a network element to the service provider is just as absurd.
Even when it comes to contracts, it can often be heard that the real selling occurs after the agreement is signed. There are always outs for the customer, even if it means paying a penalty for lost businesses from the vendor. A case in point has apparently been Cisco Systems with its metro 100G systems (please see our latest metro 100G report).
[written by Mark Lutkowitz]