By Casimer DeCusatis
One of the main topics of discussion at OFC this past year was the state of the optical components industry. While optical networks continue to grow, both in conventional telecom markets and emerging enterprise data center applications, high volumes alone aren’t enough to insure the future of the optical transceiver market. Just as many other technologies are becoming commodities (differentiated only by price) due to disruptions in their supply chains, optical transceiver vendors are coping with a variety of market forces which requires them to re-engineer their approach to delivering the components which form the backbone of our optical infrastructure.
Traditional Pricing Strategies
Traditionally, networking device manufacturers have been willing to dramatically discount their hardware (switches, routers, and multiplexers) in order to gain market share. The low entry level cost was attractive to clients in both the telecommunication and enterprise data communication sectors. Some of this revenue could be recovered later through price markups on the optical transceivers, particularly if a manufacturer supported only transceivers which had passed their internal qualification testing. This can be enforced through the vendor ID field in the IC-C interface on standard SFP form factor transceivers, and similar identification methods on other transceiver packages which allow software to determine the source of the transceivers. Although form factors such as the SFP are controlled by both formal and ad hoc industry standards, it’s possible to read the source ID field and disable equipment ports if an unsupported transceiver is used.
New Pricing Strategies for Transceivers
Marking up the cost of optical transceivers was a reasonable business strategy as long as the transceivers represented a small incremental cost compared with the rest of the network equipment. This has changed in recent years, as public and private cloud networks and the Internet have grown to encompass hundreds of thousands of nodes. Optical transceivers now represent a much larger fraction of the total system cost, which in turn has led to increased pressure to lower transceiver prices. Many large cloud and telecom service providers have embraced this direction, even if it means taking a common position with their competitors. For example, the relatively new Google Fiber division competes with vertical telecom integrators such as Verizon and AT&T, but everyone agrees that the industry as a whole would benefit from commoditized optical transceiver pricing.
Pressure on Transceiver Business Models
Since service providers like these can account for up to one-third of the total transceiver market, there is increasing pressure on transceiver vendor business models. As discussed during the OFC 2016 Market Watch , optical transceiver manufacturers often make significant investments in next generation technology such as 100 Gbit/s links. These investments are usually offset by relatively high component pricing for early adopters until the cost gradually declines in proportion to volume. But market pressures are driving down the cost for early adopters, and the accelerating pace of technology adoption is causing prices to commoditize sooner than ever before. Combine these trends with the need to support multiple form factors and legacy components for a wide range of global markets, combined with the desire to introduce new technologies faster with each successive generation, and the resulting pressure on transceiver business practices becomes enormous.
Transceiver Vendors React to the Market
Meanwhile, transceiver vendors are looking to restructure their business models by driving industry standards (to avoid having to support many different form factors) and working with industry consortiums to insure their products are a good fit for short-term adoption. Getting closer to the customer’s requirements so that new products don’t sit on the shelf is a good start. Many transceiver vendors are also pursuing innovative ways to reduce their testing costs without compromising reliability, and to take advantage of a more agile SDN network which provides other ways to insure robust data delivery beyond reliance on the underlying hardware. The outcome of this market shift remains to be seen, so be sure to follow the conversation at OFC and related OSA conferences.
Do you think optical hardware is destined for a low cost, commoditized future which compromises our current notions of reliability, or does the current push towards white boxes and virtualization actually grant some relief to the business case for optical transceivers? I’d like to know what you think, so drop me a line on Twitter (@Dr_Casimer). If you’d like to chat in person, look me up this summer at the NSF Enterprise Computing Conference, Internet 2 / NYSERNET Tech Summit, or Cisco Live. I’ll look forward to talking with you!
Posted: 8 June 2016 by
| with 0 comments