The embarrassing 400G: When will the operator backbone network be commercially available?

400G backbone network, a term that has not yet arrived and seems a bit outdated.

As early as 2013, when operators decided to skip 40G and upgrade their backbone networks directly from 10G to 100G, 400G backbone networks were on the agenda. In 2014, Wei Leping, a well-known expert in the industry, pointed out that "as long as the transmission distance and price are appropriate, 400G backbone network speed will be a more reasonable choice."

Since 2013, the three major operators have started to build 100G backbone networks on a large scale, and in a few years, they have quickly built the world's largest 100G backbone network. Since 2017, the metropolitan area network has also been gradually upgraded to 100G.

However, 400G has always been "only heard of the stairs, but no one is seen coming down", which is really embarrassing at the moment when 800G is gradually becoming popular. The noise in the industry is getting smaller and smaller. What is the reason that hinders the commercial use of 400G? How long will it take for the 400G backbone network to be implemented? At the 2021 China Optical Network Conference held recently, technical experts from China Telecom, China Mobile, Huawei, FiberHome, and Changfei conducted in-depth discussions at a roundtable forum.

China Telecom's Tang Jianjun: There is no demand at present, 400G is not cheap

Tang Jianjun, senior engineer at the Network Technology Research Institute of China Academy of Telecommunications Research, pointed out that the commercialization of 400G has been under discussion since 2017, but there is no clear conclusion as to when and what 400G technology will be commercialized, and the 400G industry chain is still immature.

On the one hand, from the demand side, there has been no urgent need to upgrade the backbone router ports to 400G in the past two years, and it is difficult to determine when the demand will arise. On the other hand, it is cost-effective. In theory, 400G is cheaper than four 100G ports, but in reality, there is no mature technical solution, the price of 400G ports is higher, and there is no advantage in power consumption and operating costs, so operators lack motivation.

Tang Jianjun pointed out that from the perspective of backbone networks, 400G is the next generation of transmission technology. Higher speeds such as 800G will take longer. Although Internet traffic has maintained rapid growth, the backbone network is flat, and the actual traffic growth is only a dozen percentage points, especially for inter-provincial backbone networks 400G, which still has a long way to go. "There is no hope for 400G to be truly deployed on a large scale in the past two years."

Wang Dong of China Mobile: 200G is a transition, 400G is not yet mature

China Mobile has established its important position in the field of optical transmission through dozens of 100G centralized procurements. Wang Dong, project manager of the Basic Network Technology Research Institute of China Mobile Research Institute, said that from 2018 to 2020, China Mobile will also deploy 200G in its backbone network. However, the most basic judgment is that 200G is a transitional option, and 400G is a generational technology.

However, 400G has not yet met the transmission distance requirements of the backbone network. China Mobile is currently working on technical solutions, but the problem cannot be solved in the short term this year.

Huawei's Bai Zhongqiang: 400G is in the research and innovation stage

Huawei is the undisputed technology and market leader in optical transmission equipment. At the forum, Bai Zhongqiang, vice president of Huawei's optical transmission field, believed that 40G/100G/200G can achieve better power consumption integration based on a set of devices. The 400G pilot is currently focused on the metropolitan area network field and has achieved good results. For the backbone network, Huawei's efforts are to make 400G have 100G transmission performance and transmission distance at the electrical layer, reuse the optical fiber and device technology of the existing network; at the optical layer, it can smoothly expand capacity and build a new 400G system.

From a historical perspective, 400G is currently in the research and innovation stage. Huawei hopes to work with operators to promote the evolution of 400G through various innovations, testing, and standard setting.

Fenghuo Ma Jun: 400G needs systematic research

FiberHome Communications is also an important supplier in the global optical transmission market. Regarding 400G, Ma Jun, a technical expert at FiberHome Communications, put forward four viewpoints.

First, network traffic is still growing, and operator networks need 400G speed and capacity. Second, the core purpose is to reduce the cost of each bit of transmission. Third, the evolution of previous generations of optical transmission technology is linear, and 400G is difficult to take into account both transmission distance and efficiency at the same time, and systematic technical research is needed, including modulation, reception, optical fiber, etc. Fourth, 400G deployment also needs to consider maintenance issues from an engineering perspective.

Overall, Ma Jun is optimistic about the large-scale application of 400G in the operator backbone network. There will be clear deployment time points and technical routes to achieve the goal of achieving the optimal cost per bit.

ZFL wavelet: It is recommended to pre-embed G.654.E optical fiber

YOFC is a leader in China's optical fiber industry and also a leader in optical fiber technology innovation. Lan Xiaobo, director of the YOFC Group Innovation Center, stressed that the 400G era will definitely become a reality. At the same time, backbone network optical fiber has been used for more than 20 years and may not meet the needs of the 400G era. Operators should plan their networks in advance and pre-bury new G.654.E optical fibers. When 400G is deployed two years later, G.654.E optical fibers can be used to obtain the best link cost.

YOFC has also made early arrangements for the more cutting-edge space-division multiplexing optical fiber, and its related product research and development is at the forefront of the industry. Lan Xiaobo believes that space-division multiplexing optical fiber is very useful in data centers because it does not need to be amplified, and only needs to be connected at both ends. However, the commercial use of 400G long-distance trunk lines is still a long way off due to factors such as optical devices, and requires industry chain manufacturers to work together to solve this problem.

From the attitudes of the technical experts of the above five representative companies, the industry can roughly see the current embarrassing situation of 400G, as well as the challenges and commercial expectations of its development in the next few years. As the moderator of the forum, Mao Qian, a well-known expert in the industry, concluded that when operators deploy 400G, they must fully consider the reuse of existing network resources to save the cost of expansion and evolution. Equipment manufacturers are cooperating with operators, and optical fiber manufacturers also have the ability to mass-produce new G.654.E optical fibers. The industry also needs to explore how to realize the productization and engineering of 400G, gain recognition from operators in terms of performance and cost, and start 400G deployment.