7 Reasons to Compute at the Edge

April 17, 2020

The “things” that make of the Internet of Things (IoT) things aren’t as inert as they might seem. Rather than merely serving as conduits for data systems or the cloud, some argue that these so-called edge devices need an intelligence boost. This is the thinking behind Hewlett Packard Enterprise’s Edgeline Systems, which provide computing power to edge devices.
Dr. Tom Bradicich, an HPE VP and GM of Servers and IoT systems, believes upgrading edge systems is critical to the future of the tech industry. Here are seven reasons why.

In a perfect world, there would be no latency or severed connections. However, in the real world, latency is still an issue, and it limits many applications. After all, what good is a smart car if latency prevents it from stopping in time to hit a wayward pedestrian? For some mission-critical functions, latency is unacceptable, making computing at the edge a necessity. Even when 5G is fully rolled out, it’s unlikely to be as fail-safe as today’s edge computing.

Transferring data from edge devices to the cloud consumes a massive amount of bandwidth. Some IT experts propose creating a separate network for IoT as such devices can drag down the system with the constant need to send data back and forth. It’s because of this that many companies can’t take on IoT’s bandwidth needs.

Companies that embrace IoT must often grapple with complex compliance issues. This can be particularly challenging as many countries have different laws and polices that govern the regional transfer of data.

Data zipping back and forth is vulnerable to breaches and attacks. As recent headlines have shown, hackers are cracking into everything from baby monitors and children’s toys to cars and aquarium thermometers.

The more bandwidth and security devices require, the higher the costs. Most companies want to save money and are constantly seeking new ways to operate more efficiently and cost-effectively.

When users collect data and send it to the cloud, some duplication is inevitable. While it might not reach 100%, if they collect 10 TB of data on the edge and then send it to the cloud, then that’s a duplication.

Data corruption
Data can corrupt on its own without any help from hackers or other ne’er-do-wells. The result? Retries, drops, and missed connections. For mission-critical applications, issues that plague edge-to-data center communications are huge deals.

To help address these seven issues, and others, HPE offers deep edge computing. While tech upgrades like 5G may help alleviate some traditional IoT computing issues, it won’t solve them all.

To learn more about HPE deep edge computing solutions, email the SYNNEX HPE Inside Sales Team.

Author: Dr. Tom Bradicich, VP, HP Fellow, Global Head of Edge and IoT CoE & Labs