Index Fragmentation

When you add new documents to the base table, the index is synchronized by adding new rows. For example, if you add the DOC 7 document with the word dog and synchronize the index, you now have:

DOG DOC1 DOC3 DOC5
DOG DOC7

Subsequent inserts, updates, or deletes also create new rows, as follows:

DOG DOC1 DOC3 DOC5
DOG DOC7
DOG DOC9
DOG DOC11

Index fragmentation occurs when you add new documents and synchronize the index. In particular, background inserts, updates, or deletes, which synchronize the index frequently, generally produce more fragmentation than batch mode synchronization.

When you perform batch processing less frequently, you reduce fragmentation because you produce longer document lists with a reduced number of rows in the index.

You can reduce index fragmentation by optimizing the index in either FULL or FAST mode with CTX_DDL.OPTIMIZE_INDEX.