Archive for the 'dba' Category

Really Big Elephants Data Warehousing with PostgreSQL

Thursday, January 10th, 2013

Presentation by Josh Berkus (PostgeSQL developer)
at MySQL User Conference 2011

Really Big Elephants Data Warehousing with PostgreSQL

Monday, October 31st, 2011

Presentation by Josh Berkus (PostgeSQL developer)
at MySQL User Conference 2011

How InnoDB performs a checkpoint at Xaprb [by Baron Schwartz {of Perconna, MySQL vendor }

Friday, May 6th, 2011

An article about MySQL InnoDB checkpoint-ing mechanism written by Perconna folks. Read comments as well, several PostgreSQL core committers have participated in the discussion.

“InnoDB’s checkpoint algorithm is not well documented. It is too co…

How InnoDB performs a checkpoint at Xaprb [by Baron Schwartz {of Perconna, MySQL vendor }

Friday, May 6th, 2011

How InnoDB handles REDO logging – MySQL Performance Blog [by Ewen Fortune]

Friday, May 6th, 2011

How MySQL InnoDB implements transaction log, aka REDO (Oracle), aka WAL (PostgreSQL)

How InnoDB handles REDO logging – MySQL Performance Blog [by Ewen Fortune]

Friday, May 6th, 2011

OTN Discussion Forums : SQL*Plus 10.2.0.1.0 does not work on … | times(NULL) = loop

Friday, May 6th, 2011

It appears to be an interesting bug in some versions of Oracle 10g, on Linux 2.6.x oracle process goes in loop doing times() system call if machine uptime < 199 days. Nice!!! / Oracle refs are bug 4612267 and 5867987 ) . The solution is to reboot.

InnoDB Flushing: Theory and solutions – MySQL Performance Blog

Friday, April 29th, 2011

blog posts that describes how MySQL InnoDB flushes data to disk (also see bgwriter, checkpoint_* parameters in PostgreSQL)

Valentine’s tech log: [PostgreSQL] Index sizes depending on the type of the field being indexed

Saturday, April 23rd, 2011

empirical data about difference in index sizes depending on the data type

Understanding the StackOverflow Database Schema – SQLServerPedia

Tuesday, April 19th, 2011

Notes about schema used by StackOverflow.com and related websites. It is very simple, only several tables used. (most likely because there are performance problems with joins)