Scaling to 40G and Beyond: The Risks and Rewards of Innovation

This article was prompted by xchange Editor in Chief Paula Bernier’s recent Q&A with Infinera CEO Jagdeep Singh.

Innovation leveraging cutting-edge technology can create unique opportunities for differentiation, but also unique risks. The rewards can be lucrative when the differentiation gained offers service operators significant TCO advantages, is sustainable and addresses a large market opportunity. But the potential risks include being on the “bleeding edge” of development, development costs, market timing and the need to establish additional proof points for new technology.

Infinera Corp. and Nortel Networks Ltd. aggressively have developed cutting-edge technology for long-haul DWDM, intending to exploit the resulting differentiation to gain market position as the market progresses from 10G to 40G and beyond. This transition invites innovation, since transmission beyond 10G involves significant increases in chromatic and polarization mode dispersion, and a variety of new modulation formats, amplification technologies and dispersion compensation techniques, receiver technologies, and analog or digital technology can be applied to overcome these impairments.

Infinera made significant market inroads at 10G with competitive cost, speed of provisioning and access to the electrical domain (and performance monitoring data) at every node. These advantages came from its Photonic Integrated Circuit (PIC) technology innovation, which consolidates the functions of multiple analog components onto digital integrated circuits.

Developing the PIC changed Infinera from being a startup in an established product category with flat growth and a lot of competition during a downturn to an established competitor with significant market share riding the next wave of revenue growth at 10G. But Infinera also took on the future risk of achieving timely PIC fabrication for native 40G and 100G. Infinera plans to achieve native 40G PIC production by year-end 2009, but will be late relative to the initial ramp of native 40G – a factor we think influenced Infinera’s strongest success being weighted more toward Tier 2 and Tier 3 operators to date. Infinera can support 40G wavelength services and router interfaces in the interim with a 4x10G solution.

The pros of this strategy include not having to deal with the extra chromatic and polarization mode dispersion encountered at 40G, enjoying 10G span design rules, fewer components, the cost advantages of the PIC prior to native 40G, and the same advantages the PIC offered at 10G when they do get native 40G. The cons are using four wavelengths on the fiber instead of one to support 40G, the associated power requirements with the 4x10 solution, and time-to-market for native 40G.

Nortel’s DPQPSK (dual polarization quadrature phase shift keying) modulation format helps overcome the effects of transmission nonlinearities, and Nortel has developed some advanced digital signal processing algorithms and is first to implement coherent receiver technology.

The pros of this include greater potential distance to 2,000km with reliable transmission at 40G, without the need for Raman amplification, regeneration or dispersion compensation equipment; and a potential leg up on development of 100G, where coherent receiver technology and more sophisticated modulation formats than DQPSK may be required. The cons are that the cost of developing this technology might make it hard for Nortel to recover the costs with 40G sales given the window of market opportunity, and could make it difficult to achieve strong margins – and proof points must be established.

Nortel’s significant investment in technology for 40G and 100G is its bid to create and leverage technology differentiation and time-to-market advantages to regain its once-dominant position in long-haul as transmission speeds increase beyond 10G.

Nortel and Infinera have taken roads less traveled in search of technology differentiation for long-haul at 40G and 100G. Nortel has raised its visibility in the long-haul market considerably in recent months as a result. Infinera will have opportunity to do so as it achieves native 40G, particularly if it can demonstrate proof points that native 100G development will be on a more aggressive road map timeline.

For competitors like Alcatel-Lucent, Ciena Corp., Huawei Technologies Co. Ltd. and Nokia Siemens Networks, the potential risk of not commercializing similar degrees of technology differentiation in long-haul is the chance that Nortel or Infinera gains market share as each exploits current and future technology innovation at 40G and beyond. The potential rewards are not risking ending up on the “bleeding edge” of innovation, having potentially lower costs than a Nortel, and time-to-market advantages in attaining native 40G and compared to Infinera.

David J. Dunphy is principal analyst, managing partner at research firm Telecom Strategy Partners.

Share this article: Email, Slashdot, Digg, Del.icio.us, Yahoo!MyWeb, Windows Live Favorites, Furl
Add this article feed to: RSS, My Yahoo, Newsgator, Bloglines