5G, Internet of Things (IoT), and growing video-based data transport place heavy pressure on carriers and data centers to upgrade their network capacity to support these data-intensive applications. In addition, recent behavioral changes brought on by the COVID-19 pandemic such as remote working, remote learning, and increased streaming for entertainment will continue well after this health crisis ends. As the explosion in the capacity demands of data-hungry applications outpaces current high-speed transport abilities, 400G is a promising new technology that supports an immediate need in fiber optics with comparatively low operational expenses (opex) and smaller footprint.
400G technology is still in its early stages, and suppliers are in a race to find the optimal specifications to meet current market needs by supporting mixed protocols, speeds, and distances. According to market researchers, data consumption will increase by more than 50% every year, and with COVID-19 the expected increase is even higher. In some hyperscale data centers, data is doubling every year.
With 100G/200G speeds, carriers and hyperscale data centers have become accustomed to interoperability issues among different vendors using different DSP-based coherent modules and forward error correction (FECs). The upcoming ecosystem of 400G DSP and optical modules will mitigate these issues, changing the landscape from customized to standards-based solutions.
Two main approaches
Two main standards dominate the market today. One is published by the Optical Internetworking Forum (OIF), 400ZR, and the second is published by the Open ROADM multisource agreement (MSA). The 400ZR Implementation Agreement recently published the OIF is designed to reduce the cost and challenges associated with high-bit-rate data center interconnect (DCI) and network flexibility. It creates a simple, footprint-optimized, and economical method for transmitting 400 Gigabit Ethernet over DCI links for short distances (typically up to 80 km). The 400ZR specification uses DWDM and higher-order modulation such as 16QAM.
The 400ZR specification also has an industry-derived variant called ZR+. The ZR+ module consumes about 15 W (more than the 400ZR specs), which allows more powerful signal processing techniques to span 200 km. The ZR+ compromises between the unique needs of the data center market and the telecom market. It offers the smallest form factor to push 400G speeds but relieves the reach limitations of 400ZR.
400G pluggable coherent modules solve some of the challenges faced by different industries and segments in the data transport market. Short distance 400G pluggable transceivers using PAM4 demand multiple lasers that cannot operate at long distances. For this reason, 400G pluggable coherent modules use coherent optics that combine phase and amplitude modulation as well as orthogonal polarization in transmission with approximately 64 Gbaud rate. This enables the 400G pluggable coherent modules to offer up to 1,000-km reach with a single laser, and as a result, far lower investment and opex. The 400ZR standard also calls for unprecedented interoperability at high speeds, which enables operators to mix equipment from multiple vendors in the same network for the first time.
Applications for 400G are not all created equal and each one has its own needs and goals. Luckily, the industry is quickly adapting to customize the market approach for each one. As there are strict power consumption limitations from the form factor of the module, the 400ZR DSP uses the OIF Concentrated FEC (CFEC) and has limited functionality; the module supports only Ethernet client interfaces with no additional features, such as encryption, tunable laser, and OTN rates. It is also not possible to connect the 400ZR over the ROADM infrastructures and metro networks of carriers. This has led the Open ROADM standard committee to define a stronger Open FEC (oFEC), which will enable carriers to expand their metro networks with 400G links or wavelengths.
400G coherent pluggable module form factors
400G coherent interfaces at the edge of networks have seen a significant reduction in power and size. They have not only become more efficient but are experiencing a boost in performance with lower complexity, a smaller footprint, and availability from several vendors supporting the same standard.
There are three main 400G pluggable modules that will dominate the market: CFP2, QSFP-DD, and OSFP. The CFP2 has the largest footprint and highest power dissipation, which provides the best optical performance and reach. The QSFP-DD has the smallest footprint and limited power package, and focuses on less demanding optical performance applications. The OSFP is somewhere between the QSFP-DD and CFP2 in size and has better power dissipation capabilities.
Due to its performance, the CFP2 400G module is best deployed in metro and long-haul applications for carriers. It supports applications such as alien wavelength and transport over multi-degree ROADM networks.
Challenging links require the use of the unique soft-decision forward error correction (SD-FEC) employed by each of the DSP vendors. The use of SD-FEC enables high performance and supports 400G transmissions of up to 1,000 km. Due to this high performance, higher power dissipation is required from the optical pluggable module. Activating the full capabilities of the DSP with SD-FEC enables 200G transmission in long-haul applications of over 2500 km and expanding long-haul 100G networks with 200G wavelengths.
Market-standard pluggable optical modules enable the use of the same modules in all DCI, metro, and long-haul applications; selecting the right module is based on the required link performance. The use of 400G MSA pluggable coherent optical modules, standard FEC modes, and the same management system eliminates the need for proprietary, bulky, non-standard single-source modules. Such use also results in lower cost per bit, provides better performance while consuming less space and power, and optimizes the link budget.
Large quantities of 400G pluggable modules are expected to be commercially available in Q1 2021 and offer much lower power consumption and prices compared to current 400G/600G bespoke modules. There will be a number of vendors from which to choose, leading to fierce competition in pricing and quality. Each of these modules requires different tradeoffs between performance, specifications, and cost, depending on the link and system requirements (see Table).
As 400G technology becomes more mature in terms of standardized form factors and the interoperability associated with 100/200/400G speeds, these issues will be resolved. Until that happens, it is important that carriers and data centers consider their 400G needs in the immediate future because the road will become much simpler over the next couple of years.
Koby Reshef is CEO at PacketLight Networks. He has more than 25 years of experience in both technical and marketing aspects of the telecommunication industry. Reshaf’s expertise covers a variety of technologies such as storage networking, optical networking, ASICs, and wireless communication solutions. Before being nominated as a CEO, he managed PacketLight's hardware design team. Prior to joining PacketLight, Reshef served as a senior R&D manager at Teledata (ADC). He has an MBA from Netanya Academic College and a BSc in Computer Science from Tel-Aviv University.