db-derby-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Mike Matrigali (JIRA)" <j...@apache.org>
Subject [jira] Created: (DERBY-2721) improve performance of index lookups of string types in territory based collated databases.
Date Tue, 29 May 2007 18:01:17 GMT
improve performance of index lookups of string types in territory based collated databases.
-------------------------------------------------------------------------------------------

                 Key: DERBY-2721
                 URL: https://issues.apache.org/jira/browse/DERBY-2721
             Project: Derby
          Issue Type: Improvement
          Components: SQL, Store
    Affects Versions: 10.3.0.0
            Reporter: Mike Matrigali
            Priority: Minor


Comparison of strings in a non unicode basic collation database is going to be slower than
comparing unicode values directly.  Indexes perform
many such compares before finding a single key in a search.  

Some suggestions for future work:
o Somehow cache  CollationKey's for the strings and use them directly for compares rather
than the actual strings.
o even better would be to use a functional index that actually stored CollationKey's rather
than the strings themselves, and then build the index on the
   CollationKeys.  Depends on some sort of functional index work.  Something like  	 DERBY-455
would build the infrastructure to make this kind of
index easy.

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.


Mime
View raw message