Re: Re: Best "how to index" docs/books

From: <l.flatz_at_bluewin.ch>
Date: Tue, 14 Jan 2020 14:18:40 +0100 (CET)
Message-ID: <1000589387.24092.1579007920015_at_bluewin.ch>





Hi Franck,
happy new year. That shows that the exact need of Kyle is necessary to know. If I take "to find optimal opportunities for adding indexes." I am using throw-away from Martin Berg. Simple. If you have a full table scan that only returns a few rows and throws away a lot, you should add an index. My scripts work like that. Regards
Lothar
----Ursprüngliche Nachricht----
Von : franck_at_pachot.net
Datum : 14/01/2020 - 11:37 (MN)
An : contact_at_soocs.de
Cc : kylelf_at_gmail.com, Oracle-L_at_freelists.org Betreff : Re: Best "how to index" docs/books  Hi Kyle,  

  To developers who want to see practical cases and examples, I recommend Markus Winand   https://use-the-index-luke.com/ website and book. Simple to understand, with examples more than theory. And specificities of DB2, MySQL, Oracle, PostgreSQL, SQL Server. Free website, and cheap book, available in English, French, German, Spanish, and Japanese.    

  Franck.    

  On Tue, Jan 14, 2020 at 9:13 AM Stefan Koehler <   contact_at_soocs.de> wrote:       

  Hello Kyle,   

 not really a book but Lothar Flatz did such a project in the past and has written an article about it in DOAG/RedStack magazine (   https://www.doag.org/formes/pubfiles/6663723/2015-01-News-Lothar-Flatz-Faktenbasierte-Indexierung-ein-Erfahrungsbericht.pdf ).      

 Not quite sure how good your German skills are for reading this article but you may also can get in touch with him directly (   https://twitter.com/LotharFlatz ) :-)      

 Best Regards   

 Stefan Koehler      

 Independent Oracle performance consultant and researcher   

 Website:
  http://www.soocs.de   

 Twitter: _at_OracleSK      

> kyle Hailey <

  kylelf_at_gmail.com> hat am 13. Januar 2020 um 21:30 geschrieben:   

>
  

> What are the best docs or books out there on methodologies for Index?
  

> The first past approach is just to look at execution plan, see where there are full tablescans scanning significant # of rows or blocks with predicate or join filters that filters out most of the rows. Then extending that index opportunity to other SQL and seeing which columns might be included in the index and what order. I like this description from a PostgreSQL extension
  https://rjuju.github.io/postgresql/2020/01/06/pg_qualstats-2-global-index-advisor.html.   

>
  

> Many years ago, Mogens Nørgaard recommend Dan Tow's book "SQL Tuning" to me for SQL tuning methodologies and to this day it has been the best example of a prescriptive , step by step process, to find the best (or near best) execution plan for a SQL statement, and ended up helping me lead a [project to implement it in software](
  http://dboptimizer.com/2011/07/08/woohoo-vst-takes-the-next-step/).   

>
  

> Would like to find something similar for indexing to find optimal opportunities for adding indexes.
  

>
  

> Feel free to say what not to read. Knowing what not to read can be important as well.
  

>
  

> Thanks
  

> Kyle
  

 --   

  http://www.freelists.org/webpage/oracle-l           

--
http://www.freelists.org/webpage/oracle-l
Received on Tue Jan 14 2020 - 14:18:40 CET

Original text of this message