How to Successfully Approach Edge IoT

As more devices become connected and the IoT continues to grow, Edge Computing or "the Edge" is coming up often in digital transformation discussions… and it should be. However, we've noticed there is a lack of resources to go to for Edge IoT information. So we’re sharing some tips on how to successfully approach Edge IoT that we’ve learned from our own experiences with digital experiments.

The Partial Approach to Edge IoT
The current buzz around Edge IoT heavily promoted by major cloud vendors such as AWS, Microsoft Azure , Google Cloud and IBM Watson IoT is the ability to connect cloud or analytics intelligence to an edge device. For example, Microsoft allows developers to build models in Azure and push those models to an IoT gateway. While that is great functionality, what happens to those models once they're on the IoT gateway? This is a very IT-centric approach, which isn't wrong; it's just not complete.

What else has to occur to get those models to do something valuable or actionable? Where do you go to see the output and result? What if you want to share those models with another edge device, person or place? How do you configure the edge device to do what you want? Now multiply this by the number of edge devices across your organization and you can see how complicated it can get and quickly see there are many other often overlooked costs associated to incorporating a single edge device. All the major cloud providers are part of our incredible ecosystem; we love them, and we partner with them heavily on this topic because there is so much more to consider. As you continue to think about Edge IoT, ask yourself these questions to avoid any nightmares.

Questions to Ask When Thinking About Edge IoT
Think about the basics

Where do we want to run our intelligence (on premise, on the actual asset, on the factory floor)?
What type of edge devices do we need (IoT gateway, smart sensor, data extraction)?
How do we get the edge device in place securely (physically)?
How do we power the edge device (and keep the power on)?

Think about the data flow

What do we want to connect the edge device to (what type of asset)?
How are we going to connect that edge device (via network, cable, sensor)?
How do we connect the edge device without changing anything to the machine (how do we connect without altering the machine and do so unobtrusively)?
What do we want to do with the data from the asset (example goals: discovery, meeting particular thresholds, failure prediction)?
Where does the data need to go (from, through, to: clouds, historians, databases, another edge device, asset, person, thing)?

Think about the people

How do we monitor and control the device and data?
How do we handle edge device management?
How do we manage access and security?
What happens operationally?
Who is the expert on the data outputs?
As we make discoveries and uncover conditions, what to do we do with the information (take control, turn something down or off, notify others)?

The Successful Approach to Edge IoT
A successful approach to Edge IoT is an edge-centric one, which means thinking about the Edge as the point of entry into your digital ecosystem. As our resident mad scientist and Field CTO of ADLINK IoT Joe Speed says, "Problems that can be solved at the edge should be solved at the edge." A successful approach to Edge IoT requires thinking about and planning for all of the questions listed above. Look at the basics of connectivity and data extraction; think about streaming data to the right people, places and things, at the right time; and think about the reason behind it all -- the people and the business.

IoT is all about partnering with the right people to create your own digital ecosystem within your company so you can affect cost reduction and revenue growth initiatives. Does Edge IoT need to be complicated? Certainly not, and it's the whole reason ADLINK IoT exists. We make IoT simple.
__________

Here at ADLINK we are committed to making IoT simple. If you're interested in learning more about Edge IoT, curious to know what it can mean for your business, or simply have a few questions and want to speak to someone knowledgeable, feel free to contact us or schedule a 1:1 Edge IoT briefing

Automotive

Defence & Aviation

Gaming

Healthcare

Industrial Automation

Networking and Communications

Retail & Logistics

Semiconductor Solution

Smart City

Test & Measurement

Railway

Robotics

Featured Solutions

5G MEC

Autonomous Vehicle

GPU Solutions

ROS 2 Solution

Smart Manufacturing

AMD-based Solutions

Ampere-based Solutions

Arm-based Solutions

Intel-based Solutions

MediaTek-based Solutions

NVIDIA-based Solutions

NXP-based Solutions

Qualcomm-based Solutions

Support

Downloads

Contact

Values

Company

Newsroom