I have argued before that edge computing is the secret to perfectly crisp fries. And that in itself should be proof enough that edge is indeed the future of cloud.
But I get it. Not everyone is easily convinced by my culinary argument. So I wrote this post to explain how maturing technologies like 5G and cloud-native make edge more reliable and easier to manage. I call them edge enablers and here are some of them:
- IoT and connected devices are unique data sources that need to be secured and registered in the cloud. Edge will reside near or on these data sources.
- Containers provide a standardized deployment environment for developers to build and package applications. Thanks to edge computing, containers can be deployed on various hardware, regardless of device capabilities, settings and configurations.
- Service and data mesh provide a way to deploy and query data and services distributed across containers and datastores across the edge. These meshes present a single interface that abstracts away the routing and management of services and data interfaces. This critical enabler makes possible bulk queries for entire populations within the edge, rather than on each device.
- Software-defined networking allows users to configure the overlay networks. It also makes it easy to customize routing and bandwidth to determines how to connect edge devices to each other and to the cloud.
- 5G makes edge deployments seamless by guaranteeing the transmission of critical control messages that manage the edge. This last-mile technology connects the edge to the internet backhaul and ensures that edge devices have the right configurations and software versions to do the right things.
- Digital twin is a critical enabler that organizes physical-to-digital and cloud-to-edge. The twin allows data and applications to be configured using domain terms around assets and production lines rather than database tables and message streams. Digital twins allow domain experts (rather than software engineers) to configure applications to sense, think, and act on the edge.
4 technologies that are making edge computing even more powerful
I’m already seeing these exciting applications in my work:
1. XR (extended reality) presents a truly immersive interface for users to collaborate or work in virtualized environments. Add edge, and you get even more detailed and interactive experiences.
For example, we are already creating immersive experiences like car buying, engineering site visits or worker safety training. Thanks to using edge, people can see new views and zoom in for unparalleled granularity.
2. Heterogenous hardware processes more data — faster and using less power. Using this specialized hardware on the edge embeds compute efficiently within physical environments and accelerates its response.
For example, we recently used Intel’s new Loihi neuromorphic chip to implement voice-powered commands in car, like “lights on,” “lights off,” or “start engine.” This extremely low-powered chip is deployed in the car in “always-on” mode, listening for your commands without draining the battery.
3. Privacy-preserving technology includes techniques and hardware that allow data to be analyzed without exposing all of its aspects. Examples include secure enclaves, homomorphic compute, federated learning, differential privacy. Data is typically encrypted when stored and when transmitted, but privacy-preserving technology protects the data even through the compute stage, making it more useable by other lines of business and partners especially when it needs to occur on the edge.
4. Robotics can be configured to act based on signals and updates at the edge. In fact, we just finished an edge implementation for robot-assisted surgery. While the surgical controls happen directly on the robot, the edge also coordinates with the cloud to determine which controls are deployed on the robot, what data is used, and what information is ultimately transmitted back to the cloud.
How edge computing will drive cloud computing
All these technologies point in one direction: Edge is the future of an extended cloud continuum. Here are some ways this may unfold very soon:
Extend AI and IoT: A good deal of today’s computing already happens on the edge in hospitals, factories and retail locations. Much of it operates on the most sensitive data and powers the most critical systems that must function reliably and safely. Edge can help drive decisions on these core systems. Any time you have an opportunity for AI and IoT tapping into these systems, there is also an opportunity for edge.
Creating these experiences requires more advanced applications, analytics and AI, best developed and managed in the cloud. This is true when edge data can and cannot be sent to the cloud, like when we might use cloud to generate synthetic data (e.g., created by computers).
<<< Start >>>
<<< End >>>
Create unique value in partner ecosystems: Controlling the edge means you control the data access at the closest point of action. Use this unique position to create differentiated services that may be used across the entire business and among partners.
Just like data center proximity is a premium in high-frequency trading, these points at the edge can also drive the same differentiation.
Consider the car: Edge computing is of interest to the manufacturer, but also to the insurance provider, utility and energy companies, and for city planning. Look for opportunities where edge offers new data and you can offer value to your partners.
These new edge-enabled data and services are consumed in the cloud, where they can combine with other enterprise data and applications. They will also be cataloged and managed centrally in the cloud.
Create differentiation with 5G, robotics, XR and connected devices: Edge computing is a must for maximizing the returns of these next-generation technologies. Their combinatorial effect enables new features like voice commands to your car and remote work via teleoperation. Edge allows the programmability and control we need to fold these capabilities into the business.
Developing these highly complex use cases requires more centralized compute cycles than ever to test how they work in the real world. For example, the robotics work I do with my team begins with a simulation in the cloud thanks to AWS Robomaker. We validate much of the robotics control system, the AI system and the fleet management in the simulation. We also test for variability in lighting conditions or form factors in completely virtualized environments proving the solution before any physical changes or purchases are made.
Edge computing is viable today
What thrills me about edge is its ability to drive so many technological leaps so quickly. It’s almost the stuff of science fiction. Which is why you probably are thinking that your company is not even remotely ready to think about edge.
I’m here to tell you that it’s practical today.
The enablers I mentioned in this post — IoT, XR, 5G, and others — allow you to start imagining now how edge can make your business run more efficiently, innovate faster and get more value from your ecosystem partnerships. In other words, find your crispy fries.