Reduce latency from 10 seconds to sub 10 milliseconds with edge computing.
A centralized compute model has a cloud data center-centric architecture.
But a distributed model does some light processing on-site and sends *relevant* data back to data centers.
In this episode of B2B Tech Talk, host Keri Roberts interviews Manager of Network Architecture Research at Panduit, Brian Kelly
, about all things edge computing.
Brain talks about:
- Trends in edge computing
- 4 environments for the edge
- What makes edge computing fast and precise
“Remote management is a huge piece of edge computing in every one of these environments because you're not going to have the human interaction that you're used to in a data center environment
.” — Brian Kelly
What is edge computing?
You know those Coke machines where you can select your own flavor add-ins? Those are actually edge applications.
Each retail location has a server that reports data to a centralized cloud at Coca-Cola headquarters.
The drink machines are sensors
They told Coke, for example, that caffeine-free Diet Coke was one of the top 5 beverages selected in the afternoons. Without that data, they would never have known that people were making that choice.
But edge computing isn’t just being used in retail. You can find it pretty much everywhere: industrial, warehousing, transportation, even agriculture.
Edge computing provides data in milliseconds
because it’s just a few meters from the end use.
It also processes data on-site to message about what’s important
—which isn’t that the temperature has stayed the same for the last 10 hours but that it has suddenly changed
4 environments for edge computing
Brian explained the different environments for edge applications and why they are important.
1.Highly protected indoor (data center)
Instead of having a mega data center in an urban setting, you could put a smaller one in Kankakee, Ill., about 65 miles south of Chicago, for example.
2.General indoor (office branch)
This environment is like a restaurant, an office building, a factory… in other words, a place that has limited cooling and very basic security.
3.Harsh indoor (warehouse)
Edge applications in harsh indoor environments have the potential for both dust and water, not very secure at all.
4.Outdoor (no security)
If the edge application is outdoors in Minnesota, it’s going to be exposed to freezes in January and 100-degree heat in July.
The point of categorizing edge computing applications by environment is to think through what supplemental hardware you need.
Brian said what you’re accustomed to in a data center may not be available in the other environments.
Where is technology going in the next year?
Brian: The shift from offices to home environments will mean increased network security for protecting data wherever the people are. “It's going to be a huge change in mindset in both IT and company strategies with employees,” he said.
Learn more about edge computing at Panduit
To join the discussion, follow us on Twitter @IngramTechSol #B2BTechTalk
Listen to this episode and more like it by subscribing to B2B Tech Talk on Spotify, Apple Podcasts, or Stitcher. Or, tune in on our website.