Currently the timesheet gadget on JIRA is running the following query to pick up the data:

SELECT ID, issueid, AUTHOR, grouplevel, rolelevel, worklogbody, CREATED, UPDATEAUTHOR, UPDATED, STARTDATE, timeworked FROM worklog WHERE (STARTDATE >=  'XXXX ) AND (STARTDATE <  XXXXX' );

since there is not index for the column startdate, this query need to make a full table scan with the condition as in the execution plan. Having an index on that column will dramatically improve the query execution time in some database flavors. This suggestion is to track general performance improvements that can be done to this table for all databases.