Source: Trihedral Engineering/AI
VTScada is built around its own enterprise historian, so every backup server in an application can be bidirectionally synchronized.

Is your SCADA history ready to go to work?

June 16, 2025
Straightforward principles for optimizing SCADA data relevance, safety and shareability with Trihedral Engineering's Chris Little

SCADA applications are responsible for far more than facilitating real-time process monitoring and alarm management. The process history they compile over time is critical to providing the data-driven insights that industry relies on when optimizing their systems to control costs, maximize uptime and increase the life of infrastructure. Modern SCADA systems must ensure data is safe, relevant and easily shareable with a company’s own team or third-party reporting solutions, business systems and artificial intelligence (AI) platforms.

Control talked to Chris Little, media relations director, Trihedral Engineering, about straightforward principles to ensure that your SCADA data is ready to go to work.

Q: What should people ask when they're setting up or updating their historian?

A: The first thing to ask is if your data is relevant? How do you ensure you're recording only what is necessary and not filling your database with noise? One simple way to do this is to use deadbands. For example, if you are measuring the level of a lake, the tiny changes caused by wind blowing across the surface are not helpful. Your SCADA software should allow you to set a deadband value so that only changes greater than this threshold get logged. Our VTScada software allows you to set deadbands at both the tag level and at the driver level, which also helps reduce network traffic. Also, look at the polling rate for your system. How often do process values get logged for each I/O point? Often, this rate is set high by default or because the developer assumed more data was better. Ask yourself if you really need to poll a particular sensor once per second. Odds are you don't. VTScada provides a rapid polling mode just for situations where you need to troubleshoot a problem. 

Q: That makes sense, but what else do we need to know?

A: Make sure your history is complete. Obviously, to capture all your process data, your SCADA system must be running. For that, you need at least one redundant server. For many for small- to medium-sized systems, people are reluctant to move to a multi-server environment because of the perceived cost and complexity. Often, this is because they don't consider the cost of a system failure. In addition to losing access to real-time monitoring and critical alarms, you lose a chunk of your history that you can't get back. Comparatively, the cost of another server and software license is negligible.

VTScada makes it easier to configure robust multi-server failover for any number of servers without writing a single line of code. We also have discounted multi-server bundles to help smaller systems benefit from Enterprise redundancy.

If you are replacing your SCADA system, another way to ensure your history is complete is to export historical data to a CSV file and import that into your new application. VTScada provides operators, third-party analytics and business systems an uninterrupted view. 

Q: Now that the data is there and relevant, how do you keep it safe?

A: The answer is to ensure your history is backed up on multiple servers in real-time. Again, this requires redundant servers with automatic synchronization. Bear in mind that if your hot backup servers are on the same desk in the same office, you have a problem. A flood, fire or loss of power at that location instantly negates the benefit of redundancy. Make sure you host at least one backup server at a different location. For critical systems, consider having two since an event serious enough to take down one location can easily take down another.