Just about every IT pundit has speculated about the promise of 5G to transform digital business and data center infrastructures. Recently, Equinix included 5G in its five predictions of IT trendsthat will dominate the digital future of enterprises in 2019.
I wholeheartedly agree. There’s enormous upside to 5G and the next evolution of wireless network architectures. I also am closely following 5G’s role in driving the creation of edge computing infrastructures with fiber-like speeds to power our IoT- and AI-connected world. As more data moves to the edge, however, there will be intensifying pressure on enterprises and data center providers to develop modular, distributed data centers that leverage increased levels of security, interconnectivity and operational efficiency.
One area that’s often overlooked—but should be on the front burner—is power. In edge data centers, power efficiency will be at a premium—and a defining factor in the success or failure of the operation. If power is overprovisioned, which has long been the norm in traditional IT environments, data center P&L will face an uphill battle. On the other hand, insufficient power allocation will come with a hefty price if surges caused by workload peaks adversely impact customer experience or business performance.
In centralized data centers, there’s a constant balancing act between managing energy supply and demand often due to overallocation, which sometimes requires “robbing Peter to pay Paul ” – taking energy allocated for one workload and using it to address a spike or peak created by another. Unfortunately, in distributed edge data centers, typically there is no surplus—no Peter or Paul—to support energy reallocation. Nor do you have as much control over an edge power source, which may not be as clean, consistent or reliable as what you’re used to in a traditional data center.
That’s where Software Defined Power (SDP) comes in. The best way to optimize power utilization in edge data centers is by combining software virtualization with AI, machine learning and edge hardware. The result is much greater visibility across workloads and more efficient energy utilization overall. This leads to higher rack density in a small footprint. Additionally, SDP can support dynamic allocation of power from the local utility or alternate energy sources, such as wind and solar, along with battery storage located at the edge.
With SDP, edge data centers benefit from much-needed intelligence to predict power usage patterns across all energy sources. For instance, the opportunity to better understand the impact of weather patterns if relying on alternate sources is crucial for maintaining consistent levels of performance. Equally important are analytics that identify power-capacity peaks under certain conditions, enabling power to be automatically reallocated in real-time, as needed for additional workloads.
By optimizing power utilization with SDP, data center providers and enterprises will be better positioned to fulfill the promise of 5G while being ready for the impending arrival of edge data centers.