The shift to hybrid work and widely distributed enterprises is forcing businesses to find new ways of working. It's also finally given edge computing its moment in the spotlight. With the increase in small branch offices and the uptick in mobile and IoT devices in use, edge computing is emerging as an effective way to process and store data across dispersed workforces.
One of the many promises of 5G is that it will revolutionize edge computing. Industry pundits have valid reasons to be excited, especially about high-band 5G, which uses millimeter wave frequencies and can provide high-speed, low-latency connection to devices on the edge of the network. But it will be many years before 5G could dethrone Wi-Fi as the de facto wireless technology for enabling connectivity at the edge.
Here are three reasons Wi-Fi will remain at the core of enterprise edge deployments, especially as the next generation, Wi-Fi 6, unlocks more efficient use of the existing radio frequency and Wi-Fi 6E opens up the 6 GHz band for enterprises to utilize.
There are different versions of the 5G spectrum implementation: low-band, mid-band and high-band. Low-band offers only marginal speed improvements over the previous cellular generations and aren't likely to drive fundamental shifts in cellular usage.
The real excitement is over high-band 5G, which uses frequencies above 20 GHz and offers fast speeds with low latency. The trade-off, however, is that high-band signals don't travel very far and are more easily disrupted by things like buildings, glass and foliage as compared to lower bands. This presents a unique problem for enterprise environments that typically span large areas, especially as workforces continue to grow more distributed. To get the full value of the speed and low latency of high-band 5G, microcell stations need to be positioned very close to one another, making network deployments time-consuming and expensive.
In contrast, Wi-Fi continues to be an affordable option because, in most cases, Wi-Fi infrastructure is already in place. Perhaps, even more importantly, because of Wi-Fi's long reign as the de facto wireless connectivity, there is a large base of devices that is only compatible with Wi-Fi. It would be prohibitively expensive for enterprises to replace them all. Combine this with the fact that Wi-Fi chipsets are much more affordable than 5G, it becomes clear Wi-Fi is here to stay. That said, if enterprises do want to upgrade over time, Wi-Fi 6 and 6E are both options, as they offer higher throughput, lower latency and better network efficiency than previous generations of Wi-Fi.
Better energy consumption
Another unique advantage that Wi-Fi 6 brings to the table in edge connectivity is a feature called Target Wake Time. This allows devices on the network to plan out communications with an access point, reducing the time they need to keep radios powered and improving the battery life of sensors and endpoint devices.
Currently, endpoints must wake up routinely, even if updates are only needed infrequently. This prohibits low-power designs that aim to reduce the power devices consume, while maintaining necessary functionality. As Target Wake Time becomes more widely available with Wi-Fi 6, enterprises will be able to decrease their power consumption, while unlocking new use cases that come with better connectivity.
This is in stark contrast to 5G, where data centers and towers are generally expected to increase energy consumption. In fact, some experts estimate that 5G could increase total network energy consumption by as much as 150% to 170% by 2026, according to a Vertiv and 451 Research survey. As the number of endpoint devices at the edge continues to explode, energy efficiency will only grow more relevant. It's best to start thinking about the role that wireless connectivity will play in controlling energy consumption, and Wi-Fi 6 can be a tool to drive down the energy footprint.
Reliability on par with 5G
One other primary selling point of 5G for future edge deployments is that it will offer the ability to slice connectivity into isolated end-to-end networks tailored to fulfill diverse requirements requested by a particular application. Each slice can have different characteristics and can be reserved for a specific use, providing guaranteed performance and low latency.
While Wi-Fi does not offer slicing. Many of the cellular slicing use cases are unique problems of a broadly shared macro network. Wi-Fi does not share those problems and, yet, it still has many mechanisms that can be used to control policy, traffic optimization and service delivery at the edge. When you couple those existing capabilities with the improved support for an increased number of devices that comes with the new OFDMA feature, it opens up Wi-Fi 6 as a solid option for connectivity at the edge, especially in the environments where devices require guaranteed bandwidth for consistent operations.
Don't count out Wi-Fi -- it will remain a critical part of connectivity at the edge
As the number of IoT devices continues to proliferate throughout the enterprise, processing and storing such large volumes of data in a centralized cloud or data center is increasingly impractical.
More network intelligence needs to move closer to the sensors, video cameras, cash registers and hundreds of other IoT devices that collect, process and create new data. While 5G is sure to unlock some exciting use cases for edge deployments, it is impractical to believe it will ever fully replace Wi-Fi when it comes to enabling connectivity at the edge.
About the author
Giacomo Bernardi is a distinguished engineer within the office of CTO at Extreme Networks. He plays a key role in setting the company's overall product strategy and roadmap and how to innovate within emerging technologies. Bernardi's research interests focus on machine learning and wireless networking. Previously, he was CTO of a privately held startup telecom service provider with a value of over $1 billion where he built one of the largest software-defined networking deployments. He got his doctorate from the University of Edinburgh in the U.K. and a Master of Science at Trinity College Dublin in Ireland.