Your AI Edge Application Needs 5G

Artificial intelligence (AI) has a massive appetite for data. And you can only satisfy that appetite — and gain the benefits AI brings — if you have a reliable, high-bandwidth network to transport data to and from your AI applications.

When thinking about AI, people often forget about the network — until it doesn’t work. We take for granted that our applications will be able to communicate with data sources and return the valuable insights we need in a timely manner. However, as more AI moves the edge, organizations are finding that they need faster speed, lower latency, and more bandwidth than their current networking options provide.

That’s where 5G comes in.

The Edge Migration

Many emerging AI use cases, like smart infrastructure, video safety monitoring, autonomous vehicles, the Industrial Internet of Things (IIoT), AI-enabled health diagnostics and others, need to process data in real time as soon as it is generated. Moving that data from the place where it was created up to the cloud for processing and then back down to the edge simply takes too long to be useful. For example, the camera on an assembly line robot might rely on AI to help it determine if a particular part falls within spec or needs to be scrapped. If processing that data is delayed even half a second, the manufacturer may encounter considerable expense from stopping and starting the line, or it might even ship a faulty part to a customer.

To deal with these real-time processing situations, AI compute capabilities are moving out towards the edge of the network. The sensors in assembly lines, farmer’s fields, oil and gas pipelines, retail stores, traffic signals and a myriad of other locations now have servers nearby, ready to process the incoming data through AI algorithms and enable an intelligent response all within the blink of an eye.

Wi-Fi vs 5G/LTE

It can be easy to try to make an apples-to-apples comparison of LTE/5G to Wi-Fi but the two technologies have different strengths and weaknesses. For instance, it’s long been assumed that Wi-Fi is less expensive than a private LTE/5G network. But it depends on the application. While Wi-Fi networks are focused on delivering decent network coverage at a low cost, private mobile networks are usually outcome-focused and driven by an operations team to support a specific use case where security, reliability, and performance are critical. 

5G uses spectrum much more efficiently than earlier cellular technologies, allowing it to make use of a wider range of bands. It also offers faster speed, up to 20 Gbps in peak performance. In addition, it supports 100 times more network traffic and offers 10 times lower latency than 4G.^[1] In applications where efficiency, reliability, and speed are important, LTE/5G will actually end up saving enterprises money. 

Better Together: AI, 5G and the Edge

Because of these advantages, leading organizations are already beginning to deploy AI edge solutions that rely on 5G networks for some of their most critical use cases.

For example, many manufacturers are using AI and 5G to support edge-based predictive maintenance applications. Downtime can be costly for these companies, but by ingesting and analyzing sensor data from their equipment, they can forecast with certainty when a particular machine is about to fail. That allows them to plan ahead and make repairs with the least amount of disruption to production.

Many of these same manufacturers are harnessing this trio of technologies to power video analytics for health and safety or quality control purposes. In many cases, they use the AI-based systems to augment the capabilities of their human staff. That reduces the amount of time employees must spend on mundane cases, allowing them to concentrate their efforts on the situations where human judgment is necessary. That in turn lowers costs while improving safety and quality records.

Similarly, in health care, many hospitals are using edge-based AI systems on 5G networks alongside human staff to help triage patients. It allows them to perform triage tasks more quickly and more equitably. By more effectively prioritizing emergency care, these organizations are able to reduce hospital stays and improve outcomes while using their resources as efficiently as possible. 

Learn More

Of course, none of these use cases would be possible without fast edge compute capabilities designed to handle AI. Dell Technologies has created Validated Designs for AI that feature the latest-generation Intel processors. These can enable up to 18 times faster performance than competing hardware. Explore more AI 5G/LTE edge solutions from Dell Technologies and Intel.

***

Intel® Technologies Move Analytics Forward

Data analytics is the key to unlocking the most value you can extract from data across your organization. To create a productive, cost-effective analytics strategy that gets results, you need high performance hardware that’s optimized to work with the software you use.

Modern data analytics spans a range of technologies, from dedicated analytics platforms and databases to deep learning and artificial intelligence (AI). Just starting out with analytics? Ready to evolve your analytics strategy or improve your data quality? There’s always room to grow, and Intel is ready to help. With a deep ecosystem of analytics technologies and partners, Intel accelerates the efforts of data scientists, analysts, and developers in every industry. Find out more about Intel advanced analytics.

---

^[1] Alexander Gillis, “What is 5G?”, TechTarget, https://www.techtarget.com/searchnetworking/definition/5G

Edge Computing