5G Among the Innovations Driving the Need for New Network Architecture

5G, cloud and edge computing, and the IoT – any one of these monumental technology developments could be considered one of the biggest networking evolutions in human history. And yet we’re experiencing all of them simultaneously. As a combined force, they are radically changing the way people communicate – and the way networks are built to support them. While each of these is unique in its own way, one thing they share is the need for a new class of network for the communications infrastructure that underpins the vision.

5G makes its mark

The analyst firm Omdia found telecom operators added 225 million 5G subscribers between Q3 2019 and Q3 2020. As of December 2020, there were 229 million 5G subscriptions globally – a 66% increase over the prior quarter. 5G is widely considered the fastest-growing mobile technology in history.

5G telecommunications operators are building new datacenters and expanding the number of access points to a massive scale with the goal of unlocking unimaginable new services based on greater bandwidth and lower latency.

Edge and cloud computing

Gartner predicts organizations will increase their investments in edge capacity so they can reduce latency and support custom security policies and personalized content delivery. Cloud and edge computing datacenter operators continue to expand to hyper scale, delivering more intelligent, localized, and autonomous computing resources for an increasing set of users, and applications.

The Internet of Things

As innovators discover the IoT’s possibilities, more “things” are added. This includes the invention of devices and sensors spanning personal use, entertainment and quality of life, to factory automation, safety, connected cities, smart power grids, new applications of machine learning and artificial intelligence, and so much more.

Changing requirements for the network

New network designs are necessary for the possibilities that these technological innovations create – and there’s too much at stake for failure to be an option. Data center designers and operators have emerged with a clear set of networking requirements that cannot be compromised:

Efficiency: The most valuable and expensive resources within the datacenter servers, the central processing units (CPUs), spend most of their time supporting the applications, services and revenue they were intended for, offloading all burdensome workloads such as networking, data and security processing.

Agility: The network must provide hardware performance at the speed of software innovation in a world of software-defined-everything. This equates to programmability that allows the hardware to evolve as fast as new networking protocols and standards do, as well as the threat landscape driven by bad actors.

Performance: It is crucial to have line-rate networking with ultra-low latency that scales from 25 gigabit Ethernet in the servers to 100 gigabit Ethernet and beyond in the network links that connect them.

Orchestration: Organizations need new methods for automation, configuration and control to orchestrate and manage so many elements at this scale.

Economics: Cost at this incredible scale cannot be forgotten. A foundational requirement for minimal costs shifts the design of networks towards open, standard and commodity off-the-shelf products.

Security: In light of so many users, devices and applications able to enter the network in many configurations, a new distributed architecture is essential.

Sustainability: It is important to make sure that the network operates as required, but in a size and power configuration that are environmentally friendly.

The role of programmable SmartNICs

These new requirements are must-haves, so organizations are moving away from networks that are large, proprietary, expensive, monolithic, vertically integrated systems – and a new method of software-defined networking has emerged.

The decline of Moore’s Law has created the need for a new heterogeneous processing architecture where expensive and burdensome workloads are offloaded from the CPUs. This model has proven to be successful in the past, with GPUs offloading video and graphics processing from the CPU. This same model is now being applied for data, network, and security processing. However, the applications underpinned by 5G, cloud, edge computing and IoT often fail on standard servers with basic NICs.

With the sunsetting of Moore’s Law, enterprises prefer SmartNICs as the method to avoid these problems and help advance datacenter computing. A SmartNIC’s primary function is to operate as a co-processor inside of the server, offloading the CPU from the burdensome tasks of network and security processing, while simultaneously accelerating applications on multiple dimensions. It uniquely meets the network design requirements for performance, agility, efficiency, security, economics, orchestration and sustainability.

Making the most of today’s possibilities

Communication between people, between things, and between people and things has been radically enabled and transformed by the IoT, cloud and edge computing, and 5G. New networks must be built to support these developments – a whole new type of network with stringent requirements for its communications infrastructure. SmartNICs make this new network architecture possible, yielding a low-cost yet high-performance result while meeting all the other requirements. SmartNICs offload burdensome processing tasks for maximum efficiency; they’re an essential part of the infrastructure enterprises need today.

Written by Jarrod Siket.