You can help by commenting or suggesting your edit directly into the transcript. We'll review any changes before posting them.
Learn about different system architectures that Ignition supports.
Video recorded using: Ignition 8.0
[00:00] Every company can use the Ignition software in unique ways depending on the needs of their facility. Companies can even have multiple facilities each with their own unique needs for the software. How the Ignition Gateway or gateways are set up in a facility is what we like to call the Ignition system architecture. The architecture gives us an idea of where Ignition sits within a facility's infrastructure as well as how it communicates with various things throughout the plant such as PLCs, databases, and even the clients. What's important to understand is that there are not correct architectures and incorrect architectures, and you certainly don't have to follow any of these examples we have here. These examples can be thought of more as a guide to help you plan out what you need to do within your facility. To start, one of the simplest architectures that we can think of is what we call the standard architecture, which you can see in the image here. Typically, you would have an Ignition server sitting somewhere within the facility communicating with one or more databases and also communicating with one or more PLCs. It can then relay this data to any number of clients running throughout the facility. This simple architecture is where most users start when they first install Ignition at their facility. Of course, there are variations on this such as adding redundancy or placing the Ignition server and/or database server within a cloud-based environment. As your Ignition use grows, you may find that you need to adopt a new architecture to fit your facility's growing needs. With growing systems, we typically see two types of growth. The first is a growth in density of data and operations at the facility. While Ignition may be unlimited, we typically find that users run into a limitation of the hardware as their facility grows. This is what we call a scaled-out architecture where the load on the Ignition system has been scaled out over multiple Ignition gateways instead of pushing all the data through one central system. You'll notice that as the number of devices in the facility grew, we split the Ignition system into a front-end gateway and multiple back-end or tag and I/O gateways. The tag and I/O gateways communicate solely with the devices and can then push this data to the front-end gateway which supplies it to the clients. This has the benefit of spreading the communications load out to multiple points in the network instead of having everything communicate through one location. It should also be noted that as you scale out even more, the cost doesn't increase dramatically. Because the back-end gateways are only used for communicating with various devices, they don't need any visualization system modules like Vision or Perspective but simply need the device driver modules. Similarly, because the front-end gateway is only concerned with the visualization systems, you would have those types of modules on that gateway, but you wouldn't need things like the device drivers. Of course, you can add variation to this architecture as well. Here you can see we have multiple front-end gateways behind a load balancer so that as new clients come online, the load balancer automatically distributes the load amongst all of the front-end gateways. The other common type of growth that we see is a growth of geography where there may be multiple facilities spread out over a wide geography. Typically, with these types of architectures, you would want a central Ignition gateway which helps to manage all of the other facilities' gateways and facilitate communications between all of them. You'll then notice that each site has an installation to fit its needs. The remote site SCADA on the left and local site SCADA on the right have architectures very similar to a standard architecture. The local devices facility has something similar to a back-end or tag and I/O gateway that's pulling the tag data from the devices and passing it on to the central Ignition gateway. And then the remote field devices have either an Edge gateway or MQTT-enabled devices placed next to them, and those push data back to the central Ignition gateway as well. With every site sharing data back to the central Ignition gateway, you now have the ability to compare all of the data between all of your different sites. One thing to keep in mind is that even with this larger and more complex architecture, you can still plan out an individual site's architecture as well, scaling out that site as needed.