5G and The Evolution of Data Centers

neuCentrIX - 03/02/2021 10:00

The whole world is anticipating the global availability of 5G, the 5th and latest generation mobile network. By mid 2020, 5G was deployed in 38 countries, and according to a prediction by the Swedish telecoms giant Ericsson, 5G internet will cover up to 65% of the world's population by the end of 2025.

 

Compared to its predecessor, the 5G technology is meant to deliver higher data speed, lower latency, larger network capacity, better reliability and availability, and a more uniform experience to users. With all these features, 5G is expected to bring advancements to IoT applications, machine learning, and artificial intelligence (AI) — improving people’s productivity, safety, and quality of life. It will also take the mobile ecosystem into a new territory and significantly impact innovations in all industries, especially the vital ones like communications, healthcare, transportation, energy, manufacturing, and agriculture. To be able to perform well, this powerful technology requires the right infrastructure — a change to the typical IT elements and architecture which are currently supporting today’s 3G and 4G networks. So, how are data centers changing to keep their roles in the 5G-enabled world?

 

C-RAN Architecture Adoption

Today’s 5G networks are supported by C-RAN (Centralized or Cloud Radio Access Network). A C-RAN is a type of RAN (the part of a telecommunications system that connects individual devices to other parts of a network through radio connections) which has a centralized control and processing station. To support the 5G technology, data centers and cloud service providers are transitioning their sites to a C-RAN architecture and functioning as C-RAN hubs. Data centers with a C-RAN architecture will be able to produce higher spectrum efficiency, achieve faster speeds, create more scalable and flexible networks, and support a larger number of mobile users and wireless standards.

 

The Birth of Smaller Data Centers

In its development, 5G is likely to co-evolve with edge computing as both working in tandem to exploit the best features of each other. Edge computing is a way to minimize distance which is one major factor of latency. In edge computing, data isn’t sent to a centralized data center and/or cloud to be processed; instead, the network edge analyzes, processes, and stores the data based on content and service level requirements by the end users. To support it, some computing capability needs to be placed out at this edge in the form of smaller data centers which are distributed based on a strategic geographical plan and connected by a network.

 

Infrastructure for AI and Machine Learning

In a world empowered by 5G and the Internet of Things, the importance of AI and machine learning is clear. Neither is the role of the data center in making it happen. To produce the algorithms needed to drive AI and machine learning, massive data processing and storage capacity is required. Adding to that, AI and machine learning applications demand high-speed and high-capacity networks, with an advanced layer of switches connected to a large number of servers which are all working on the same problem. With the right technology and architecture, data centers, hyperscale and local micro, will come as a support to these applications.

 

To conclude, data centers require some major changes to accommodate and enable 5G, but they will certainly still have a major role to play due to their ability to deliver low-cost, high-volume computing and storage. This role further evolves to become a part of a larger 5G and edge ecosystems through smaller scale data centers which are distributed at the edge and equipped with the right technology.