A cloud is an IT environment that abstracts, pools, and shares IT resources across a network. An edge is a computing location at the edge of a network, along with the hardware and software at those physical locations. Cloud computing is the act of running workloads within clouds, while edge computing is the act of running workloads on edge devices.
Edge devices can contribute to a cloud, if the storage and computing capabilities provided by those devices at the endpoints of a network are abstracted, pooled, and shared across a network—essentially becoming part of a larger cloud infrastructure.
Edge computing is not part of a cloud. What makes edge computing so useful is that it is purposefully separate from clouds and cloud computing.
Here’s how we see it:
- Clouds are places where data can be stored or applications can run. They are software-defined environments created by datacenters or server farms.
- Edges are also places where data is collected. They are physical environments made up of hardware outside a datacenter.
- Cloud computing is an act; the act of running workloads in a cloud.
- Edge computing is also an act; the act of running workloads on edge devices.
An edge (location) is not the same thing as edge computing (action). Collecting data at the edge of a network and transferring it to a cloud with minimal (if any) modification is not edge computing—it’s just networking.
But, if that data is collected and processed at the edge, then it’s edge computing.
Edge computing is separate from clouds for 2 main reasons:
- Time sensitivity. The rate at which a decision needs to be made doesn’t allow for the lag that would normally take place as data is collected by an edge device, transferred to a central cloud without modification, and then processed before a decision is sent back to the edge device for execution.
- Data volume. The sheer volume of data collected is too much to send—unaltered—to a cloud.
They all can be connected. But they don’t have to be connected.
Clouds can exist without the Internet of Things (IoT) or edge devices. IoT and edge can exist without clouds. IoT can exist without edge devices or edge computing. IoT devices may connect to an edge or a cloud. Some edge devices connect to a cloud or private datacenter, others edge devices only connect to similarly central locations intermittently, and others never connect to anything—at all. Ever.
But edge computing, when used as part of manufacturing, mining, processing, or shipping operations rarely exists without IoT. That’s because IoT devices—everyday physical objects that collect and transfer data or dictate actions like controlling switches, locks, motors, or robots—are the sources and destinations that edge devices process and activate without relying on a central location or cloud.
- Home automation is generally an IoT exercise. Your phone and smart home devices (light bulbs, thermostats, and outlets) are all IoT devices, because they simply send data—and execution decisions—back and forth (sometimes through a cloud). Neither your phone nor smart devices are processing the data they collect.
- Satellite imagery—like the kind being used on the International Space Station (ISS)— is an edge computing exercise. Edge devices physically located on the ISS are running containerized analytical code as a single-node Red Hat® OpenShift® cluster that connects to IBM Cloud on Earth. Only images that are worth transfering are sent down to the ground. Edge computing is a necessary step here because the sheer volume of data collected is too much to send to an Earth-based cloud.
5G refers to the fifth generation of telecommunications networks, representing upgrades in bandwidth and latency. 5G is a transport mechanism that enhances the capabilities of cloud computing and edge computing—but 5G is not the edge, an edge device, or edge computing. Mobile computing is also not the same thing as edge computing. Said another way, your smartphone is (usually) not an edge device.
Because our recommended approach to enterprise technology doesn’t change depending on how or where you roll it out: Everything should just work—everywhere. In the datacenter. Across clouds. At the edge.
This is the basic premise of an open hybrid cloud strategy. An open hybrid cloud strategy lets you run applications on servers in datacenters or on edge devices across multiple clouds without having to rebuild those apps, retrain people, or maintain disparate environments. And we have a broad ecosystem of certified partners, including popular public cloud service providers like IBM Cloud, Google Cloud, Amazon Web Services (AWS), and Microsoft Azure.