Comments on: High Performance MySQL, Second Edition: Schema Optimization and Indexing http://www.xaprb.com/blog/2007/10/14/high-performance-mysql-second-edition-schema-optimization-and-indexing/ Stay curious! Wed, 09 Jul 2008 03:15:11 +0000 http://wordpress.org/?v=2.2.2 By: David Rasch http://www.xaprb.com/blog/2007/10/14/high-performance-mysql-second-edition-schema-optimization-and-indexing/#comment-13538 David Rasch Tue, 16 Oct 2007 01:33:13 +0000 http://www.xaprb.com/blog/2007/10/14/high-performance-mysql-second-edition-schema-optimization-and-indexing/#comment-13538 This might be a great place to cover the process of designing a schema including a good list of: 1. what questions you intend to ask the schema 2. what you know about the growth of the given tables (seems like this barebones template might fit well in the case study) It might also be worth mentioning the storage of IP addresses as a LONGINT as this seems to be an atypical encoding in practice. I love the comments about "smaller is better", but please don't forget the pros/cons of static and dynamic row lengths. Lastly, but not least, resist the urge to have an explicit primary key if there really isn't one (not so helfpul in InnoDB, but it can save lots of space on MyISAM). This might be a great place to cover the process of designing a schema including a good list of:
1. what questions you intend to ask the schema
2. what you know about the growth of the given tables
(seems like this barebones template might fit well in the case study)

It might also be worth mentioning the storage of IP addresses as a LONGINT as this seems to be an atypical encoding in practice.

I love the comments about “smaller is better”, but please don’t forget the pros/cons of static and dynamic row lengths.

Lastly, but not least, resist the urge to have an explicit primary key if there really isn’t one (not so helfpul in InnoDB, but it can save lots of space on MyISAM).

]]> By: Xaprb http://www.xaprb.com/blog/2007/10/14/high-performance-mysql-second-edition-schema-optimization-and-indexing/#comment-13537 Xaprb Mon, 15 Oct 2007 22:32:40 +0000 http://www.xaprb.com/blog/2007/10/14/high-performance-mysql-second-edition-schema-optimization-and-indexing/#comment-13537 Partitioning is discussed in several chapters, yes. Thanks for asking about it. Partitioning is discussed in several chapters, yes. Thanks for asking about it.

]]> By: psteve http://www.xaprb.com/blog/2007/10/14/high-performance-mysql-second-edition-schema-optimization-and-indexing/#comment-13536 psteve Mon, 15 Oct 2007 20:54:21 +0000 http://www.xaprb.com/blog/2007/10/14/high-performance-mysql-second-edition-schema-optimization-and-indexing/#comment-13536 Jay Pipes' presentation at http://jpipes.com/presentations/performance-coding/performance-coding-soup-to-nuts.pdf had some very interesting things to say about vertical partitioning (and horizontal partitioning as well). Are you going to be discussing partitioning? Vertical partitioning seems relevant to this chapter. Jay Pipes’ presentation at http://jpipes.com/presentations/performance-coding/performance-coding-soup-to-nuts.pdf had some very interesting things to say about vertical partitioning (and horizontal partitioning as well). Are you going to be discussing partitioning? Vertical partitioning seems relevant to this chapter.

]]> By: Xaprb http://www.xaprb.com/blog/2007/10/14/high-performance-mysql-second-edition-schema-optimization-and-indexing/#comment-13530 Xaprb Mon, 15 Oct 2007 11:15:15 +0000 http://www.xaprb.com/blog/2007/10/14/high-performance-mysql-second-edition-schema-optimization-and-indexing/#comment-13530 Right, we are mentioning limitations and restrictions throughout as they apply. We have so much more detail in this edition -- I'd estimate at least ten times as much information -- it would be impossible to simply add detail to the first book and get a usable book. That's actually one of the biggest challenges with the book. Figuring out how to make a brain-dump from four people into a readable narrative and usable technical reference is hard. Right, we are mentioning limitations and restrictions throughout as they apply. We have so much more detail in this edition — I’d estimate at least ten times as much information — it would be impossible to simply add detail to the first book and get a usable book.

That’s actually one of the biggest challenges with the book. Figuring out how to make a brain-dump from four people into a readable narrative and usable technical reference is hard.

]]> By: david http://www.xaprb.com/blog/2007/10/14/high-performance-mysql-second-edition-schema-optimization-and-indexing/#comment-13527 david Mon, 15 Oct 2007 05:38:46 +0000 http://www.xaprb.com/blog/2007/10/14/high-performance-mysql-second-edition-schema-optimization-and-indexing/#comment-13527 How are you going to address the differences between storage engines, e.g. on index types? The current version of the book only adds notes to the end of chapters, which don't really go into detail. I would like to read about difference as I'm reading through the chapter, so that I know right away that feature x cannot be used with innodb. How are you going to address the differences between storage engines, e.g. on index types?

The current version of the book only adds notes to the end of chapters, which don’t really go into detail. I would like to read about difference as I’m reading through the chapter, so that I know right away that feature x cannot be used with innodb.

]]>