Row set scan already increases the performance by factors if querying more than 10% of the table rows. It does not necessarily a full table scan (only if the first and the last table row is part of the row set), the IntervalScanner setups a row interval including the first and the last row of the row set, which is always ordered. During the scan the row set is always reduced by the already returned rows and the following ranges working on the reduced set.
Unfortunately the current design does not leave much room for adding additional scanner data without changing the client ScanSpec, which is passed one by one to the create_scanner method. Well I can adjust the row set scan to use the existing row_intervals, which is definitely more effort on IntervalScanner side and on MergeScanner side as well. -Andy -- You received this message because you are subscribed to the Google Groups "Hypertable Development" group. To post to this group, send email to [email protected]. To unsubscribe from this group, send email to [email protected]. For more options, visit this group at http://groups.google.com/group/hypertable-dev?hl=en.
