At first blush, it seems illogical for Infinera, a supplier, which has been fighting for its survival, to be promoting a revolutionary change in the aggregation and access portions of the optical marketplace. However, fibeReality now believes the company chose to amend its “Tellabsation” counternarrative because manufacturing its own proprietary 800-gig DSP daughter card (using subcarriers), with similar functionality to the AC1200 from Acacia Communications (planned to become part of Cisco Systems) in order to compete with Ciena’s WaveLogic 5 Extreme, is deemed critical to the future success of the Groove platform as well as for the current leadership to maintain at least a minimum amount of credibility. While nobody should bet the farm on XR Optics, given the challenging hurdles that need to be overcome, the effort would not only make the transition away from being a PIC-based company less abrupt, but more vitally, it could possibly attract funding, which could be applied for the necessary, internal high-end DSP development. In our opinion, co-founder David Welch, who has remained comfortable and contented since the extremely handsome rewards from the bubble days, felt at ease in first bucking the industry trend by initially pushing PICs for long-haul applications (albeit, hardly offering any benefits).
So, Welch is now the most ideal individual to be the front person for this new scheme, as he certainly believes that he has something to prove in coming out of the “wilderness” – and would be interested in at least saving some face. Undoubtedly, he views the XR Optics concept as kind of a hybrid of the Groove with the Cloud Xpress, and the latter can provide a talking point regarding the supposed value in the investment in PICs.
Adding to our confidence that such an otherwise counterintuitive messaging strategy will likely be effective is that we are aware that board member, Greg Dougherty, the former Oclaro, CEO, is working very closely with Infinera’s CEO, Tom Fallon, and the former wants the latter to definitely succeed. Moreover, although we still assert that there will be a relatively limited number of applications for 800G in the foreseeable future, it would presumably be a showstopper in some cases for the Groove to lack this higher data rate per lambda.
Whether the approximately $50 million to fund the DSP project is obtained externally or internally, our intelligence points to a real seriousness of purpose in accomplishing the objective. The engineers at Infinera are pretty confident that it can be pulled off.
To get a better sense of how dominant the Groove has become at Infinera, before the acquisition of Coriant, Lumentum Holdings only sold a relatively few modules to the system house. Afterwards, a lot more devices were shipped by the components player to Infinera, particularly the ACOs, which may partially account for the leadership at Lumentum being initially flat-footed on the demand potential for that particular gear.
It should be noted that with the understanding that the XR Optics is not the priority in terms of the R&D push, it will certainly make it that much harder to pull it off. Yet, the biggest problem will be in convincing established, conservative-minded operators, like Comcast and AT&T, to make a networking change in such a radically different direction.
Is it reasonable to expect the major MSOs to just suddenly abandon all of the work that has been done in utilizing the CableLabs standard for a coherent access infrastructure in favor of this point-to-multipoint paradigm? What about all of the complications involving lasers, which are overlooked by Welch in his presentations with the proposed architecture?
For our fast-growing, totally separate, quick update blog, which is exclusively on fibeReality’s LinkedIn page, please follow us here.
As always, fibeReality does not recommend any securities, and this writer does not invest in any companies being analyzed by us.
[written by Mark Lutkowitz]