The journey to the 400/800 g has been open Which route will you choose?(1)
As new technologies continue to drive the demand for higher bandwidth, coupled with the fact that the path to 400G is already in place, owners and operators of data centers of all types and sizes need to understand current standards-based options and how to implement deployments, while keeping a close eye on developments in 800G and higher speeds. This is beneficial to help them identify which route to 400G best meets the needs of their organizations and how to optimize their infrastructure to fully leverage the potential of existing technologies to enable digital transformation and sustain business growth.
Growing Cloud Applications
Digital transformation of enterprise businesses and adoption of emerging applications are driving the demand for 400G applications in cloud data centers. Growing demand for work-from-home, online transactions, and video traffic is significantly driving cloud adoption, according to Synergy Research
Group's findings show that cloud spending grew by 25 percent in the first quarter of 2020 alone. The results show a 25% increase in data center hardware spending on public clouds and a 3% decrease in the enterprise and non-cloud service provider markets.
Even in the face of a recession caused by the new crown epidemic, cloud spending is expected to continue to grow, with Cisco predicting that 94 percent of work will be performed in some form of cloud environment by the end of 2021, and Gartner forecasting that spending on public cloud services will reach nearly $700 billion by 2025. As cloud adoption increases, so does the number of hyper scale data centers, which reached 541 in the second quarter of 2020 alone.
Advances in coding schemes, transceivers and optical communication technologies are also key drivers for 400G adoption, providing the technical and economic feasibility to enable high rate applications. One of the major advances is the development of the PAM4 (four-stage pulse amplitude modulation) coding scheme, which has twice the bit rate of the previous NRZ (no return to zero) coding scheme, supports rates of 25, 50 and 100 Gb/s per channel, enables a more efficient migration path of 25G-50G-100G-200G-400G-800G, and reduces the amount of the amount of infrastructure needed to support the application. For example, a 40GBASE-SR4 40G application using NRZ encoding at 10Gb/s per channel requires 8 cores of fiber, 4 transmit and 4 receive, all at 10Gb/s, while 200G and 400G applications can be supported on 8 cores of fiber using PAM4 encoding at 50Gb/s and 100Gb/s per channel.
Transceiver technology is also advancing, with pluggable transceiver modules evolving from NRZ-encoded 10Gb/s per channel to PAM4-encoded 50Gb/s per channel. the latest QSFP-DD and OSFP interface pluggable transceivers are 8-channel 50Gb/s transceivers that can support 400G applications. Several QSFPDD and OSFP interface transceivers are available on the market today for a variety of multimode and singlemode applications. These pluggable transceivers with fiber optic cabling and connectors provide the flexibility to support different transmission speeds and leverage existing hardware.