Re: GSoC Query
От | Gokulakannan Somasundaram |
---|---|
Тема | Re: GSoC Query |
Дата | |
Msg-id | 9362e74e1003290342i645d27a7sa7fd21b417432b56@mail.gmail.com обсуждение исходный текст |
Ответ на | Re: GSoC Query (Greg Smith <greg@2ndquadrant.com>) |
Список | pgsql-hackers |
<div class="gmail_quote"><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt0pt 0.8ex; padding-left: 1ex;"><div class="im"><br /><br /><blockquote class="gmail_quote" style="border-left: 1px solidrgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;"> Similarly using the no. of select hits on a tablewe can check that if maximum no. of times it is on a non-index field we can index on that field to make select faster.<br/></blockquote><br /></div> It's impractical to figure out where indexes should go at without simulating what theoptimizer would then do with them against a sample set of queries. You can't do anything useful just with basic statisticsabout the tables.<br /><br /> I would recommend <a href="http://msdn.microsoft.com/en-us/library/aa226167%28SQL.70%29.aspx" target="_blank">http://msdn.microsoft.com/en-us/library/aa226167(SQL.70).aspx</a>as a good, practical introduction to thetopic of what it takes to figure out where indexes go at, from someone who came up with a reasonable solution to thatproblem. You can find a list of the underlying research they cite (and an idea what has been done since then) at <ahref="http://portal.acm.org/citation.cfm?id=673646" target="_blank">http://portal.acm.org/citation.cfm?id=673646</a><divclass="im"><br /></div></blockquote></div><br />Evenif you have devised a way to find the appropriate set of indexes, just have a index adviser, which would advise a setof indexes for a set of queries and let the DBA and the application user take the final call, after looking at them..<br/><br />Gokul.<br />
В списке pgsql-hackers по дате отправления: