Synchronization

The TMS Synchronization batch job runs on individual TMS instances. It checks for changes in the TMS repository that impact the Omissions and Source Terms tables and:

• Updates records in those tables

• Marks the Update flag for the same records so that they are reprocessed during the next data exchange between TMS and the external system (Batch Validation in Oracle Clinical)

If you have TMS integrated fully or partially with an external source data system, you must run Synchronization. TMS Synchronization is included in Oracle Clinical Batch Validation when the two systems are integrated.

If you are using TMS as a look-up system only, without integration to an external source data system (such as Oracle Clinical), you do not need to run Synchronization.

This section includes:

• Scheduling Frequency
  
• Data Processed During Synchronization
  
• Synchronization Processing Order
  
• Running Synchronization
  

Scheduling Frequency

Oracle recommends that you run Synchronization at least five times during the business day at each site. You should also run Synchronization before and after batch jobs that exchange information between TMS and an external system. When TMS is integrated with Oracle Clinical, this happens automatically during Batch Validation.

Data Processed During Synchronization

The Omissions table contains source terms that Autoclassification could not classify. The Source Terms table includes all locally collected source terms and the classification (VTA ID or dict_content_id) for each one. When an omission is classified manually, the Synchronization job moves the source term from the Omissions table to the Source Terms table, associated with its classification. If a VTA is changed at any point, Synchronization changes the VTA ID in the Source Terms table.

The following types of changes require synchronization with the Source Terms table:

• Classifying source term omissions

• Declassifying and reclassifying verbatim terms

• Approving and unapproving VTAs

• Loading a new dictionary (to mark the dictionary for future processing)

• Loading a new dictionary version (to mark the dictionary version and its source terms and omissions for future processing)

• Changes to Informative Notes that are derived to the external system

In addition, if TMS is integrated with an external system, the following changes to source terms in the external system must be synchronized in TMS:

• Definitional changes such as question definition changes or mapping a study or project to a different domain in Oracle Clinical

• Deletion of source terms in the source system (Oracle Clinical responses)

Synchronization runs on data imported using the TMS DSI feature; because the systems are not integrated, the importing instance owns the data.

Synchronization Processing Order

Synchronization always includes two phases, pre-synchronization and post-synchronization.

• Pre-synchronization: Pre-synchronization runs first. It updates records as outlined Data Processed During Synchronization and sets the Update flag to Y so that records will be updated in the external system during the next data exchange (Batch Validation in Oracle Clinical).

• Post-synchronization: Post-synchronization runs last. It refreshes materialized views if the TMS_CONFIGURATION reference codelist value SYNCREFRESHMV is set to Y (see SYNCREFRESHMV).

External System Integration: If TMS is integrated with an external source data system, additional processing occurs between these two phases during data exchange between the two systems. The data exchange job must:

• Run full autocode on all source terms that have been changed since the last data exchange and on all records in the Source Terms and Omissions tables whose Update flag is checked

• Reset the Update flag on all processed records to N

Oracle Clinical Batch Validation is predefined to do both.

The batch processing unit (a study in Oracle Clinical) is called an X Area in TMS. The following indexes use the X Area to give fast access to source terms and omissions that need to be processed:

• TMS_SOURCE_TERMS_BI3

• TMS_VT_OMISSIONS_BI1

Both indexes include the following columns: NONUNIQUE X_AREA, NONUNIQUE DEF_INTEGRATION_KEY, NONUNIQUE UPDATE_FLAG.

Note:

Although Batch Validation runs on only a single study, or TMS X Area, Synchronization always runs on all TMS data, even when run during Batch Validation.

Running Synchronization

To run Synchronization:

1. From the Definition menu, select Jobs, then Synchronize Dictionary Data. The Synchronization batch job window appears. There are no parameters to enter.
2. Use the runner icons in the toolbar to save and retrieve parameter sets as necessary, submit the job, and monitor its progress.

See Reports and Batch Jobs for more information.