Migrate PI Batch data to event frames

Starting with PI Server 2015, you can migrate your PI Batch Database and PI Batch Subsystem data to event frames. Event frames help you capture, compare, and analyze important processes or business events over a repeatable time period. They can track many kinds of events, batches being just one type. Event frames provide wider search options, and greater flexibility in composition than using batches. For more information about event frames, see the PI Server topic Structure of event frames.

Because you will need different client applications to manage batch data in event frames than you have been using with the Batch Database, do not migrate to event frames until the solutions you need for event frames are all in place. Also, once you start the migration, the Batch Database becomes read only, and you will not be able to revert to it.

We do not recommend that you run the PI Batch Database and use event frames in parallel for the same batch data. We envision that you will migrate all batch data at one time from the PI Batch Database. An exception is if you decide to create event frames to support non-batch data (such as for monitoring equipment downtime) at the same time as you create batches in the PI Batch Database.

Note: We recommend that, unless you have non-critical migration requirements or a very simple environment, you test the migration of your batch data to event frames and the cutover to event frame data sources and consumers.

Allow approximately 90 minutes per million event frames for the migration, bearing in mind that this time will vary greatly depending on network, hardware and other factors.