5G is all about connecting the world, and distributed edge clouds will become a big part in how this happens. There are several reasons for this – speed, efficiency and cost. Speed has become critical in how data is delivered, analyzed, processed and consumed. Traditionally, data is analyzed within a core data center. This means that every bit of data – even the packets that aren’t required for analysis - must be routed from the user, to the central data center, and back to the user. All this traveling takes time, wastes resources, and can result in a poor user experience.
When 5G powers millions of devices at home, around the community and in workspaces, decisions will need to be made in fractions of a second to as to where to send a device’s traffic to maintain uptime and provide a superior user experience. When there are large volumes of ‘perishable’ data that need to be analyzed, it becomes much more efficient if all those data packets can be processed where they are being generated - instead of traveling hundreds or thousands of miles to a data center and back. At the same time, the capacity of these machines to interact and share data create additional traffic burdens. And for new, interactive 5G services like connected cars and virtual and augmented reality, It’s not only inefficient, it’s a deal-breaker.
So, carriers are looking for ways to bring as much of this as possible closer to where the data is being generated – In other words, to the edge. While some analysis will still need to happen at the core, IDC’s FutureScape for IoT report believes that 40% of initial IoT data analysis will occur at the edge by 2022. Think of edge analytics as the first layer of analysis – separating out what doesn’t need to be analyzed and only sending back what does. But what constitutes the ‘edge,’ and how much can we actually do there? That question brings us to the first of our five steps:
- Create the Right Architecture
The ‘edge’ is at or close to where the data is being generated – typically near the end user. Edge cloud extends the traditional private and public cloud to the edge of the network, closer to the last mile. But analyzing data at the edge requires the right network equipment. Its success depends upon a cloud-based architecture, which provides greater elasticity and fault tolerance. This means organizations will need to invest in gateways to aggregate and analyze edge data, but now is the time to make these investments.
- Collect and Combine
The ability to instantly collect and combine internal and external data sources, such as social media and other web-based information - at the edge - means the level of decision-making can become more complex. More data = better decisions.
- Aggregate and Analyze
Once data is collected it must be aggregated and analyzed. With edge computing, more of this work can be performed more efficiently where the data is being generated. This allows organizations to obtain real-time results from IoT devices for visualizations and insights, while dramatically reducing the amount of data which travels back to the core datacenter. This helps reduce network congestion and keep costs in check – as well as provides faster response times for mission critical services.
- Respond and Deliver
Edge technology enables hyper-personalized, interactive experiences where content providers can react and engage with customers securely and with sub-millisecond latency, creating a hyper-personalized experience that’s dictated and controlled by the end user. This opens the door to creating more effective ways of reaching consumers, providing timely offers with higher acceptance rates and a better customer experience.
- Measure, Assure and Predict
Distributed edge clouds will transform content delivery services, creating new partnerships, services and complex revenue share arrangements that will need to be measured, monitored, predicted and assured. With 5G and the IoT, the number of digital transactions will explode – potentially congesting and crippling traditional network architecture and opening new avenues for risk and fraud. By keeping these transactions automated and at the edge, much of the real-time processing and data prediction can be completed where and when it happens.
If you are interested in learning more about the benefits of edge-based cloud analytics, join me for our upcoming webinar: Assuring Edge Cloud Success, September 12, 2019 at 14:00 BST.