Log Buffer #55: a Carnival of the Vanities for DBAs

Ladies and gentlemen, boys and girls, cats, dogs, goats, cows, furry objects, Kevin Clossen, and whoever else may have accidentally stumbled here: the great blogosphere may have appeared to gradually slow. But fear no more, the tides are changing! You may now stop pinching yourself and trying to find your doctor’s phone number because this is not a dream and you are not having strange hallucinations. This is NOT a test of the emergency broadcast system; this IS the real thing.

YES…

Congratulation on becoming a genuine reader of the 55th edition of the OFFICIAL, INTERNATIONAL, EXTRA-TERRESTRIAL, INTERGALACTIC, ASTROLOGICAL, DECIDUOUS

Log Buffer

First and foremost, a quick thanks to that great wonder of wonders, that marvelous man of men – Dave Edwards at Pythian. He’s invited me to host this week’s log buffer and to “have fun.” I would also like to point out that this is the first log buffer I’ve posted – and it actually wasn’t quite as easy as I thought it would be. There is in fact quite a bit of activity in the database blogging world and it’s impossible to capture it all in a short summary! Bravo to all the great editors who have put these together in the past. And make sure to visit the aggregators (such as orana, orablogs, and planetmysql) and read all the posts I’ve left out! And now for the carnival…

Conferences and Open Source Databases

Although Robin Harris cursorily mentioned two other conferences (on scalability) and Stewart Smith mentioned linux.conf.au, the elephant in the room is clearly OSCON in Portland.

There were a number of pre-conference events. Robert Treat gave the low-down on the PostgreSQL party and Joshua Drake did the same. And Matt Asay from CNET talked about Tim O’Reilly’s keynote at the Ubuntu Live Conference just ahead of OSCON – where Jay Pipes snapped a pic of MySQL CEO Mårten Mickos’ competitive side – who also delivered a keynote at Ubuntu. Raven Zachary reviews Martin’s take on MySQL adoption of the GPLv3.

Then OSCON officially started. Lots of people posted updates. Jonathan Cheyer noted the demos of new solidDB for MySQL HA options. The usual MySQL vs PostgreSQL comparisons were on… Mike Kruckenberg wrote about those PostgreSQL benchmarks after a session with Josh Berkus and Peter Zaitsev did the same thing. Mark Atwood and Peter Zaitsev both posted general updates from the conference. Kaj Arnö discussed the “State of the Dolphin” and Arjen Lentz talked about MySQL stability.

And then there was the big power outage in San Francisco on Tuesday: indeed I myself was without email when a major datacenter called 365main didn’t kick in the backup generators. Resultantly, Artur Bergman discussed keeping data safe and Chip Turner drew the connection to MySQL backup integrity. Which reminds me that Matt Asay recently posted about Zmanda, a solid Amanda-based backup solution for MySQL.

There was also a high signal-to-noise ratio for PostgreSQL-based EnterpriseDB this week. Lewis Cunningham suggested that “Oracle should be terrified” and discussed compatibility. The buzz may have started with a webinar last week which got Peter K interested and made Kevin Closson investigate.

Oracle

However Kevin actually spent most of that post talking about Oracle Licensing for “development systems” and for “self-education” – a very hot topic on the oracle-l list this week (subject: vmware & Oracle). This also prompted Niall Litchfield to explain and Howard Rogers to counter. But Justin Kestelyn cleared up the definitions on the Forum Thread – and even updated the downloads page! Oracle may have updated their downloads page in less time than it took 365main to start their backup generators…

In other Oracle news, Howard Rogers and Dominic Delmolino continue that great tradition of posting SQL puzzles – asking for a problem that can only be solved with PL/SQL (not SQL) and a SQL-only query to generate 100 prime numbers. Make sure to read the comments; there are some interesting solutions!

Let’s see, what else? Kevin continued his Manly Man series and Lewis posted tips about feature combinations, subquery factoring and NLS-related index problems. Mark Rittman cautions us not to leap mindlessly into 11g and Tim Hall celebrates and laments his invitation to be an Oracle Ace Director. Charles Schultz continues his valiant struggle to defeat the mighty RAC and Dan Norris discusses Oracle’s latest acquisition. And there’s lots more… I just can’t fit it all in here!

SQL Server, DB2 and IMS

Last but not least there were a few posts on the Microsoft and IBM fronts. Decipher wrote about an important new feature coming on SQL Server 2008: a Resource Governor. Kalen Delaney pointed out an impressive tool for SQL Server called SQL Internals Viewer. And Chen Shapira pointed out 5 important ways that SQL Server is different from Oracle.

Didn’t see much about DB2 this week other than a post from Craig Mullins about index compression. If you use Informix then you’ll want to make sure to catch the webcast covering performance features in Cheetah. Guy Bowerman also gave a quick description of the release festivities in his post about the new version of IDS.

And that wraps it up for this edition of Log Buffer! As a closing thought I would like to say that I have discovered a fool-proof way to keep spam-robots from leaving junk comments on your blog. Basically you just add this advanced captcha.

Everyone have a great week and we’ll see you next Friday for Log Buffer #56!