CFP2-ACO Modules: Lingering Cash Cow

Avatar photoPublished on
Blog 1537563909 225

In fibeReality’s last blog article, “Oclaro: Waiting for the Other Shoe to Drop,” we analyzed some of the probable consequences for the vendor after the closing of the purchase by Lumentum. In this piece, we will concentrate on the future performance of Oclaro’s ACO transceivers, its most successful product family in recent times. Despite the company’s revenues on coherent technology as a whole remaining fairly consistent, there are current challenges on profitability. One prominent analyst on the Street expressed to us the view that post-200G, CFP2-ACO, there is nothing left in the Oclaro story. Before the announcement of the Lumentum purchase, we agreed with this perspective in that there was never a convincing narrative for a second act, and that the plan presented in moving forward was based principally on market hype (unusual for the leadership), as it was all about the earlier preparation for an exit. Nonetheless, concerning the current situation, we believe its ACOs still have some legs to continue to support cash flow for at least a couple of years (not to mention its most recent record sales quarter for this equipment). Specifically, the slow movement to 400G, which Oclaro seemingly counted on earlier this year, given the distance limitations (expressed indirectly) with the speed, should extend the life of these components, as starting with that data rate, the shift to DCOs will start to occur. Apparently, there are even new opportunities for ACOs, as part of an evolutionary shift to DCOs. We will also take a look at the ramifications of Oclaro getting blindsided (along with other large players) by the downfall of LR4 in having to compete with CWDM4.

Unsurprisingly, Oclaro is still selling a lot of ACOs to Cisco Systems, but sales of the devices to Ciena have somewhat dropped off. Furthermore, Nokia Networks is a significant customer of Oclaro’s ACOs. While the supplier has some other new customers for the product, they are more limited in quantities.

There have also been some noteworthy market developments in Asia involving Oclaro’s ACOs, although the amount of business is uncertain. For example, in working with a DSP vendor, NTT Electronics, the partner built a plug-in module, which was inserted in a switch from Edgecore Networks.

So, Edgecore has slots in the chasses for ACOs, and then ultimately for DCOs. Evidently, the customer could not figure out how to integrate the ACO because it did not have the capability to do the DSP-optics integration.

Moreover, Oclaro is supporting Edgecore now with its software suite. This solution facilitates working transmission, after the tiny plug-in DSP line card is inserted.

Of course, there is the Facebook TIP initiative as well, related to whites boxes for telecommunications, which involves ACOs. Although Oclaro is also involved in some standard solutions with the project, we have never been a big fan of the concept.

Still, the clear message at Oclaro is that in order to get the margins back up in the future, CFP2-DCO and the new 400G products will have to be hits. At the same time, there remains a healthy, justified skepticism about the rate of change in the marketplace.

Turning to other components, it is no secret that Oclaro is close to exiting the QSFP28 module business, or at least the 100G CWDM is possible. The company is still selling inventory, and continues to see some LR4 demand. Such gear is being sold by some competitors for $250 or less per unit. LR4 itself was overkill on long reach for many places in which it was being used, with large suppliers, including Oclaro, reaping the rewards, as there was very limited competition, before the volume in the market fell out.

Thus, Oclaro shifted its focus to the chip sales. However, the trick is to figure out how to ramp up sufficient revenue to support the fixed costs of the fabs. At least it is a better strategy than just staying in a losing business.  

As always, fibeReality does not recommend any securities, and this writer does not invest in any companies being analyzed by us.

To follow us on our totally separate, quick update company blog, which is exclusively on fibeReality’s LinkedIn page, please click here.

[written by Mark Lutkowitz]