400 Gigabit Ethernet (400GbE) is a Realty Now

400 Gigabit Ethernet (400GbE) is a Realty Now

Welcome to the world of terabit transport. 400GbE is a great step towards terabit per port transport capabilities and given the availability of COTS (Common Off The Shelf) hardware, service providers, data centers and enterprises are now able to build infrastructure to create enhance user experience and accommodate growing demand for bandwidth. 400GbE roll out has begun and it is expected to accelerate rapidly.

Service Provider

With the roll out of 5G and 5G based services, demand for bandwidth at low cost per bit would be essential for service providers. Good news is that COTS hardware now offers state of the art SOC (System on Chip) allowing complex processing and offloads for multiservice traffic.  For example, Broadcom® Qumran family of ASICs offer delivery of 5G transport a reality allowing service providers to push 10GbE to 100GbE at the cell site and aggregate the traffic at Edge aggregation.

Figure 1. Typical deployment of 400G edge aggregation for telecom networks.
Image

Today, many ODM (Original Device Manufacturer) and OEMs (Original Equipment Vendors) are offering 400GbE 2RU to 6 RU form factor switches ideal for edge aggregation. These devices are robust enough for building MPLS core with 5 million LSPs (Label-switched Paths) or more. Similarly core routing functions is also supported in those COTS hardware making it easier for service provider to design service-oriented flat network architecture enabling faster data transport, integration of more endpoints and enhance user experience.

Data Center

As data centers transition to faster more scalable network and relatively flat architectures from traditional hierarchical model, such as the 2-tier leaf-spine, the need for higher bandwidth with efficient connectivity becomes imperative. The leaf-spine architecture requires massive interconnects as each leaf switch fans-out to every spine switch, maximizing connectivity between servers. Big data, artificial intelligence, and deep learning functions in data centers all consume high bandwidth, forcing high-end data centers to quickly move to next generation interconnects operating at higher data rates. Today, many data centers started 100GbE and 400GBE deployments for their spine/leaf architecture.

Figure 2. Typical deployment of 400G Switch in super spine for Data Center Clos architecture.
Image

The leaf and spine architecture enable large scale data centers to create self-contained cluster of POD providing easier integration and on demand service upgrades without impacting entire data centers. For such deployments, 400GBE super spine can be useful to integrate multiple self-contained POD clusters.

Enterprise Applications for 400GbE

The increase adoption of BYOD (Bring your own Device), IOTs and bandwidth crunching applications are placing added strain on enterprise networks. IT organizations have been consistently struggling to increase network capacity to meet these mounting throughput requirements. Building a network of 10GbE is simply not enough given today’s demand on bandwidth. Enterprises now can benefit from upgrading networks with 100GbE for access transport and 400GbE for aggregation.

Figure 3. Typical deployment of 400G Switch in Enterprise Networks architecture.
Image

400GbE aggregator switch now makes it easier to scale Enterprise networks while enabling site integrations much easier.

Blog Categories

Recent Posts

Connect

Subscribe