Recently, 400G Ethernet has become a hot topic in the communication industry. It is both an outlet and a challenge. Network equipment manufacturers have launched 400G switches, and optical communication products have also launched 400G optical modules. In this article, ETU-LINK is about the impact of the emergence of 400G switches and 400G optical modules on data centers.

First of all, let's look at the 400G optical modules, which has the same transmission rate of 400G. Compared with the four 100G optical modules, the total power consumption of the 400G optical module is lower, and the number of optical modules and fibers required is less. The package type of the 400G optical module is OSFP or QSFP-DD. In comparison, the QSFP-DD package size is smaller (similar to the QSFP28 package of the traditional 100G optical module), which is more suitable for data center applications; the OSFP package size is slightly larger, Since it can provide more power consumption, it is more suitable for telecom applications.

The 400G switch is not just higher bandwidth, but it also has more functions, such as: ultra-high-speed policies, segmentation and whitelisting; wire-speed data collection and system telemetry, enabling real-time data packets, data flows and events Visibility; intelligent caching for big data or storage and machine learning workloads with dynamic prioritization of packets; minimize latency and prioritize business-critical traffic on demand, and so on.

The emergence of 400G switches and optical modules can help data centers upgrade from 100G to 400G. Although 400G switches are an upgrade of bandwidth, at the same time, from hardware bandwidth, deployment, investment protection to SDN solutions, software programmability, this It's an overall improvement.