On Mon, May 14, 2012 at 03:15:09PM +0200, Markus Pilman wrote:
- Do you really records_in_range() implementation like in (*) Yes.
- Are estimates of 10 and 20 rows close to reality? Not at all... The thing is: we can implement records_in_range not as a function with constant complexity and since a full table scan is not faster than a index range lookup in our storage engine (even if the whole table is contained in the range), we wanted to force MySQL to make always an index lookup. If I understand everything correctly, my error in thinking was, that I thought MySQL uses the information returned by records_in_range only to chose between index lookup and fulltable scan (since on a traditional disk architecture a full table scan will be faster than an index scan which scans the whole table). But this is obviously not the case.
As far as I understand you, you're correct. Information returned by records_in_range() is used in multiple places: 1. It is used to determine whether to make range scan or a full table scan. 2. It affects the order in which tables are joined. 3. (maybe there is something else it affects also) I think you should be able to force MySQL into doing index lookups instead of table scans by returing appropriate costs from handler->keyread_time() and handler->read_time() as opposed to scan_time().
So we will have to implemented this records_in_range method. Do I understand this correct, that a logarithmic (with respect to the table size) run time will be ok for this method (InnoDB seems to walk two times through the index if I get the comments in the code correct)?
I think logarithmic time should be ok. Note that records_in_range() can be called multiple times if you have multiple ranges. For example, for query like SELECT * FROM tbl WHERE t.key < 11 OR t.key IN (15, 17, 19, 22) you will get these calls: records_in_range(t.key < 11) records_in_range(t.key=15) records_in_range(t.key=17) records_in_range(t.key=19) records_in_range(t.key=22) If you prefer to get all ranges at once, check out handler::multi_range_read_info_const() in sql/multi_range_read.cc - you can implement that function and get access to a disjoint-and-sorted list of ranges. (however, records_in_range() will still need to be implemented because there are certain cases where records_in_range() is called directly, not through multi_range_read_info_const()). BR Sergei -- Sergei Petrunia, Software Developer Monty Program AB, http://askmonty.org Blog: http://s.petrunia.net/blog