Scope

This section describes the functions of AF Server and Data Archive, and how they are supported by AVEVA PI System Connector 3 (PSC3).

Replication options

PSC3 offers three source PI System replication options per data source instance. The connector supports different replication options for each data source instance, and multiple data source instances for the same source PI System.

Note that a PI Point is replicated only once, regardless of how many times it is referenced in the AF or PI Data Archive and regardless of which replication option you select.

Note: The PI Point Queries only applies to Replicate Data Archive only option and not to any other replication setting type.

• Replicate AF and Data Archive

  This option replicates the selected AF database or root element, along with PI points referenced by the AF data reference attributes.

  If you define PI Point Queries, it does not apply to Replicate AF and Data Archive. It exclusively works with Replicate Data Archive only.

• Replicate AF only

  This option replicates the selected AF database or root element without PI points referenced by AF data reference attributes.

• Replicate Data Archive only

  This option replicates PI points from a selected source Data Archive that matches the PI point query defined in the data source configuration.

AF Server

AF stores asset framework objects, such as elements and templates, in AF databases.

PSC3 facilitates copying the entire AF database, or copying a single element branch of an AF database.

Data Archive

Data Archive is the cornerstone of the PI System and your data infrastructure, enabling high-performance data storage and retrieval by client software.

PSC3 facilitates:

• Copying PI points referenced in the data reference configuration strings of AF objects associated with "Replicate AF and Data Archive" data source configurations. The data reference configuration string of AF objects may reference one or more Data Archives.

• Copying PI points associated with "Replicate Data Archive only" data source configurations that match the defined Include PI Points that match this query option.

The following diagram provides an example PSC3 configuration.

PI System Connector Architectural Diagram

PSC3 maintains a connection to the source AF and Data Archives to ensure changes occurring on the source AF and Data Archives are reflected to the respective destinations. This includes AF object additions, changes and removals. For PI points, point additions and data changes are reflected to the destination Data Archive.

This functionality relies on a discovery algorithm that periodically (every two minutes unless the PI Point Discovery Scan Rate has been modified) examines the source servers and checks for changes. PI point archive data is copied using a backfill algorithm on start and then begins monitoring current archive data events.

PI point snapshot data is monitored for data events from source PI points that have compression turned off, as shown by the PI point attribute compressing=0. It is also tracked if the PI point is selected by a Snapshot query in the Snapshot Query table found in the Overview tab.

AVEVA PI System Connector 3 general description

When replication options Replicate AF and Data Archive or Replicate AF only are configured, PSC3 initially exports an AF database (or source element) with all referenced objects (with some restrictions detailed later on). This includes the elements, element templates, categories, enumeration sets, internal tables, UOM objects and reference types.

For a single source element, only the referenced objects of the source element and the children of the referenced objects are exported. Children that are not parented by elements in the source element branch are re-parented to an element in the destination element branch.

Note: UOM objects are always copied even if a Collect Data from a Specific Branch is checked.

During AF processing using the Replicate AF and Data Archive data source configuration, any referenced PI point found in an attribute or attribute template’s data reference is included in the list of replicated PI points. This includes all PI points regardless of the source data archive. PI points that do not match any query defined in Replicate Data Archive only data source configurations will not be included in the list of replicated PI points.

The following figure shows the data flow from multiple PI Systems to multiple destination PI Systems. Observe that streams from the Connector are strictly one-way to the destinations, which means that a connector cannot obtain any information from either destination AF or Data Archive.

PI System Connector Diagram

PI Connectors communicate with PI System Connector Brokers over encrypted AMQP connections, which allows them to be on different computers.