By Ross Saunders
The growth fundamentals of our industry are strong. The multiplicative drivers of increasing broadband penetration, demand for streaming video services and Broadband on the Go will all help drive growth of the Internet. So why are we still so nervous?
The simple answer, of course, is that the competition is insanely tough. Increased competition from system vendors with superior cost structures is driving new cost points in the industry that ripple throughout the whole optical food chain. The ability to adapt quickly to changing market conditions and technology will separate the weak from the strong. Many vendors with high cost structures will need to make a strategic retreat from the more commoditized segments of the market since they do not have the cost structure to compete. This will be painful and require down-sizing to focus on narrower product areas where technological advantage can be leveraged to extract maximum value, particularly in core technologies such as chip foundries, ASICs and software.
From a market and technology standpoint, it is clear that 100G is emerging as a key focus area and demand from service providers is high. Of course, as 100G takes-off it is at the expense of lower speed 10G and 40G interfaces, so it’s clear we had better catch this bus. There are reasons to believe that 100G may stick around as a popular data rate for quite some time, perhaps even as long as 10G, due to: (i) 100GE port availability on routers/switches (at 40G we only had Packet over SONET/SDH, which was pricey); (ii) the fact that 1TE ports on routers may take a while to develop; (iii) the fact that Shannon’s Limit makes 100G DP-QPSK with SD FEC a good solution for 1,500-2,000km type “sweet-spot” distance. [ Higher spectral efficiency > 100G (Mary QAM) will mean shorter distance, in general; (iv) Electro-optics are affordable at 25-32Gbaud rate. This makes DP-QPSK affordable, but 4λx25G design options can also be attractive. Beyond this bandwidth the RF electronics/electro-optics become exotic (i.e, expensive).
Rapid product development execution is as important as trying to read the market and develop new products to intersect what the customers may likely want a couple of years out. As technology changes so often in our industry, a project delay of a year or so can kill a business case. Most development initiatives have real options built-in, such as aggressive cost reduction programs that kick-in and are financed from the initial release of a new product. Missing the timing window on new product introductions initially results in a loss of market share, but can also have ongoing negative consequences to the health of the product throughout its lifecycle. Speedy development, whilst managing/mitigating technology risks, is an area for innovation in itself.
Competing in the optical industry will be exciting, challenging and scary for the foreseeable future. The healthy creative destruction in our industry necessary to deliver radically lower cost/bit/km communications technology may cause a few more grey hairs and sleepless nights for those of us in the industry, but will grease the wheels of an ever more powerful Internet, affordable by the masses. A worthy cause.
By Ross Saunders, General Manager, Product Strategy, Opnext Subsystems Business
Posted: 24 February 2012 by
Ross Saunders
| with 0 comments