Run time processing

Run time processing refers to the day-to-day operations of theMaster Data Engine, once the initial bulk cross match (BXM) phase has been completed.

While cross matching is still performed as new sources are brought online, or as records from existing sources are added to the Master Data Engine database, it is done incrementally (by using incremental cross match [IXM]) as opposed to in bulk cross match (BXM).

The typical performance footprint of run time processing is:

• CPU loads are relatively light, but dependent on customer scaling requirements
• Application server memory demands are relatively light
• Database CPU loads are heavier than application server loads, in search-intensive environments
• Database memory demands are greater than for bulk processing
• Storage demands shift from sequential operation to random