Overview of IBM FileNet CFS for Content Integrator

The primary components in an IBM® FileNet® Content Federation Services for Content Integrator system are the exporter, the importer, the Federator database, and the Federation Administration application.

The following diagram illustrates the content federation architecture:

Based on the rules that you define in the Federator Administration application, the exporter builds a list of documents to federate from the external repositories that are accessed through IBM Content Integrator. The exporter generates federation requests for these documents and adds the requests to the queue in the federation database. The importer processes the requests and federates the specified documents into a Content Platform Engine object store.

The Federator database contains a set of tables that are used exclusively by the exporter, a set of tables that are used exclusively by the importer, and a set of queue tables that define the interface between the exporter and importer. The Federator database can be on any accessible server.

The exporter performs the following tasks:

• Runs rules against an external repository. Documents that match the criteria are queued for federation. Rules are run in order of priority, as indicated by the location of the rule within the list of rules.
• Populates a table in the federator database with documents or version series update requests that match the criteria that are specified by the rule. The table contains the metadata (in XML format) of the documents to be federated. Each document has a unique identifier, which is the same as the URN that is stored in Content Platform Engine.
• Maps properties from the external repository to Content Platform Engine properties by using the mapping rules that were defined in the IBM Content Integrator Data Map Designer application.
• Configures the batch size for a federation run according to the supplied values. The batch size establishes the number of items that the exporter can process in one run, and the query block size determines the size of each query chunk. For more information, see Configuring chunking and sweeping.

You can configure multiple exporters to enhance performance. You can install another instance of the exporter by extracting the CFS_HOME\fed_exporter.zip file to a different computer.

The importer performs the following tasks:

• Retrieves batch and content information from the Federator database.
• Tracks the data that was imported and determines whether a request is a first-time import or a reimport request (that is, property or version updates to a previously exported document).
• Creates federated content in Content Platform Engine and adds the URN link to the document or document version in the external repository.
• Maintains a mapping of the external version series data to the Content Platform Engine version series in the Federator database.

The Federator database stores the following information:

• The metadata of the documents that are selected for federation from the external repository.
• Federation status and time stamp information for batches and for individual items.
• Content federation configuration items, including logging levels and the content integration server authentication credentials.
• Content federation rule information, including the constructed queries that form part of a rule, the target Content Platform Engine folders that are required in each rule, and records declaration options.
  Restriction: IBM FileNet CFS for Content Integrator does not support annotations.