If Benjamin Franklin were alive and working in the telecommunications industry, he would say that there are three certainties: Death, taxes and the need to transmit data more quickly. Cable operators and their vendors can't do a thing about the first two, but are taking their best shot at the third.
Currently, cable operators and other service providers generally traffic data at a maximum speed of 100 Gigabits per second (Gbps) on a single wavelength (or "lambda.") That data throughput is thought to be sufficient to handle the demands for a few years. However, the industry already is thinking ahead to the next step, which is 400 Gbps transport.
They actually are thinking about it in two ways. Currently, trials by operators focus on using quadrature amplitude modulation (QAM) on "super channels" comprised of slower connections. This will enable the effective throughput to be 400 Gbps or more. Demonstrations of this approach had been mounted by Bright House Networks and Fujitsu - in a trial that was announced last week - as well as Comcast (NASDAQ:CMCSA) working with Ciena (NYSE:CIEN) and Shaw partnering with Alcatel-Lucent (NYSE:ALU).
The timeframe for the commercialization of this approach is coming into focus. Systems will be deployed in some number next year, with more significant rollouts likely in 2016, said Helen Xenos, the director of product and technology marketing for Ciena.
The most recent of the trials involves Bright House and Fujitsu. The test covered 249 kilometers and in a three-span network between Tampa and Orlando, FL, said Bill Beasley, Fujitsu's principal solutions architect for packet optical networking.
The technology from Fujitsu and the other vendors shows great promise. In the Bright House trial, throughput reached 800 Gbps. John Dickenson, Bright House's senior director of network architecture and strategy, said peak speeds were not reached. "We could have gone to a terabit [per second], but we didn't have enough equipment," he said.
The technology challenge is significant. Advances in add/drop multiplexers and development of reconfigurable optical add/drop multiplexers (ROADMs) are necessary, said Ron Kline, the principal analyst for network infrastructure at Ovum. Another tricky piece is Nyquist shaping, in which the waveform is adjusted in order to comfortably squeeze them into the super channels.
The longer term goal is to provide a "native" 400 Gbps system in which there is no need to combine discrete lower speed channels. Though the total speeds will be equal, the advantage of a native 400G system is that the spectral efficiency - the number of wavelengths dedicated to the process and the equipment at headends and other locales - is greater than that of the combined super channels. It simply is a far more efficient way of doing things. In the big scheme of things, perhaps in the distant future native 400 Gbps channels themselves can be made into super channels to support terabit-per-second transmission schemes.
It won't happen for a while, however. Experts say the industry currently is awaiting various standards-setting bodies to do their work. Ed Neslusan, the vice president of sales and marketing for MultiPhy - manufacturers of one of the key elements in the optical ecosystem that eventually will support 400 Gbps - said three bodies are most directly involved.
The Institute of Electrical and Electronics Engineers will dictate the specifics about the electrical signaling, the International Telecommunication Union will determine the framing standards that control the format of the data as it traverses the network and the Optical Internetworking Forum will lay out the specifics about the optical modules. "Those things are yet to be determined," Neslusan said. "They must be before products are brought to market."
The telecommunications industry's move to 400 Gbps - first with super channels comprised of slower services cobbled together and pushed fast by the use of 16-QAM and later by native 400 Gbps links - may seem futuristic today. But if the almost universal belief that data traffic will continue to expand proves accurate, 400 Gbps services will become necessary in a surprisingly short time.
The path that the vendors are following seems clear - and quickly becoming a reality. Keven Drury, an optical market leader at Alcatel-Lucent - which demonstrated 400 Gbps capabilities with Shaw Communications almost a year ago - said France Telecom's Orange has a 400 Gbps link operational between Paris and Lyon. It uses Alcatel-Lucent equipment, he said.
Kline, speaking specifically to the Bright House/Fujitsu trial, summed up where the industry is in general: "This is a step in the right direction," he said. "You need to be able to demonstrate that you can offer the technology. That's the first thing. Then you have to get it to market. There are a lot of moving parts, and they are all really difficult."
Carl Weinschenk is the Senior Editor of Broadband Technology Report. Contact him at [email protected].