There are four types of Transaction Groups, and they each handle data a little differently:
All Transaction Groups can execute at a set rate or on a schedule. A trigger can be used to determine when the group should record. You can use Ignition's expression language in the trigger to allow complex logic to determine when logging occurs, making precision execution easy.
Historical Groups quickly and easily store data from the plant floor into any kind of SQL database. Items from any or all devices can be included in the same group, just drag a few tags over and start the group running. Ignition will log the data until you tell it to stop.
Standard Groups are the most flexible group. They are capable of not only storing OPC values in the database, but can also write database values to OPC addresses or synchronize data changes between both the database and PLC. With this group you can create true realtime value tables in the database, and allow anything that can talk to the database to push values to a PLC. This is often used to create Recipe systems where the recipe values are stored in the database, and a user can select a recipe to write all your settings directly to Tags. Changing recipes is as easy as changing a tag value or selecting a name.
Transfer large amounts of data very efficiently with the Block Group. This groups allows you to send whole arrays of data to and from the database. It works just like the Standard group, but on a much larger scale.
The Stored Procedure Group allows you to use PLC data as inputs and outputs for your existing Stored Procedures. With the Stored Procedure Group, your IT department can have control over how data is entered and returned from the database.
In Distributed systems, PLCs can be spread out over great distances to remote sites. Collecting and centralizing data from each can be difficult and time consuming. To combat this problem, Transaction Groups are used as the cornerstone of our Hub and Spoke architecture. Historical Groups can be applied locally to each PLC for a minimal cost, and forward all data into a single, central, database.