Welcome to the 63rd edition of Log Buffer, the weekly review of database blogs. For those of you reading this who are not regular readers of my blog, my name is Craig S. Mullins and I’m a data management strategist for NEON Enterprise Software, Inc. It is an honor to be allowed to write another one of these carnivals. I wrote Log Buffer #8 way back in September of 2006. So a little more than a year later, here is my second stab at log buffering…
Last week a new blog written by superstars in the database community made waves with its first entry by Michael Stonebraker. Unfortunately, that entry was more of a shill for Stonbraker’s new product than anything else. Even more unfortunately, it looks like that will be the long-term status of The Database Column blog. I say this because the next three posts were also related to the column database technology that Stonebraker is touting: two on compression in column databases 1 2 and another from Stonebraker on the failure of OODB and why column databases will succeed.
Now don’t get me wrong, I enjoyed all of the material posted to this blog. But it is disingenuously named. If it were renamed The Column Database Blog, then its content would match its name and all would be well. Unless additional topics are introduced or the blog’s subheading (A multi-author blog on database technology and innovation) is revised, the future does not look too bright for this blog. I mean, c’mon, if it is a blog on column database systems, just call it that, okay? If not, then change the topic once in a while.
And while we’re on this topic, Curt Monash over at DBMS2, took the approach of just calling a spade a spade, by choosing to analyze the content of The Database Column as if it were a blog on Stonebraker’s company and products. Also in a later posting this week, Monash shared some additional criticism of Stonebraker’s columnar database argument.
Okay, so let's dig into the topical blogs - - that is, those that focus on specific DBMS technology - - and we'll start with my favorite, DB2.
DB2
Willie Favero opines this week about the new reserved words in DB2 Version 9 in his Getting the Most Out of DB2 for z/OS blog. He raises the concern that new DB2 reserved words may already be in use in your existing applications, and that potentially can cause problems.
Evidently the new reserved words in DB2 9 must be causing trouble because Willie wasn’t the only one to blog about them this week. Troy Coleman also wrote about them in his DB2utor blog. Troy shows some examples of problems you might run into with reserved words and offers some guidance.
Robert Catterall, of Robert’s Blog, shared his thoughts on data clustering in DB2, along with some advice on how to deal with unclustered data.
Chris Eaton promotes the upcoming IBM Information on Demand conference in his blog, An Expert’s Guide to DB2 Technology. The conference is scheduled for the week of October 14-19, 2007 at the Mandalay Bay Hotel in Las Vegas, Nevada. I'll be there and if you want to learn all about what's new in DB2-land, you should be, too.
If you are using SAP with DB2 be sure to click on over to Omer Brandis’ blog to participate in his system performance poll.
Antonio Cangiano in his blog, Zen and the Art of Ruby Programming, lets the world know that a Python driver for accessing DB2 from a Mac is currently undergoing internal testing at IBM. The driver is similar to the ones available for Ruby and PHP, so if you are knowledgeable with the Ruby driver, you’ll be able to use the Python one in no time.
Susan Visser updated readers of her Build your Skill on DB2 blog on the latest IBM DB2 and data-related redbooks that are available. Redbooks are free manuals provided by IBM on a variety of technical topics. Susan informs us that redbooks are now available converting from Oracle to DB2, DB2’s pureXML implementation, and working with IBM Records Manager.
And I’d be remiss if I didn’t point out my on-going series on DB2 9 for z/OS features over at the DB2portal blog. The last couple of entries there cover improvements to the COPY utility, changes made to MODIFY RECOVERY and specifying different characteristics to your COPY utilities with TEMPLATE switching.
Oracle
And it was busy within the realm of the Oracle blogosphere this week, too. First up, we had the Decipher Info Sys blog discussing a small, but very helpful improvement to the DBMS_OUTPUT package -- as of 10g R2 the line limit has been extended from 255 bytes to 32,767 bytes.
And the always informative and entertaining Tom Kyte regaled his readers with a treatise on tuning with sql_trace=true. Tom offers an in-depth treatment where he walks through an example that meticulously clarifies what he is posting about. Nicely done!
In Jeremy Schneider’s Oracle Resources and Ramblings we learn about Cache Buffers Chains and Latch Spelunking. In this informative post Jeremy walks through the results of analyzing latching in a RAC system along with some interesting results.
Jonathon Lewis shows us how to respond to the dreaded ORA-01410 (invalid rowid) error in The Oracle Scratchpad.
Then we have Justin Kestelyn, in the OTN TechBlog, spilling the beans on the Unconference that will be running in conjunction with the upcoming Oracle Open World conference. I’d never heard of an unconference before, but it seems like an interesting idea. If you want to learn more about it you’ll have to click on over to Justin’s blog.
In An Expert’s Guide to Oracle Technology, Oracle ACE Lewis Cunningham provides guidance on Converting Between XML and Oracle Objects.
If you are interested in Oracle 11g and security issues then you should check out Pete Finnigan’s Oracle Security Blog. This week Pete investigates issues with the new password hash algorithm.
In other 11g coverage, Tyler Muth shows us the new syntax to use for SQL pivoting and contrasts it with the pre-11g method which required a lot more coding.
Laurent Schneider’s blog offers up an expression for digging numbers out of character strings which can come in quite handy if you need to extract numbers out of varchar2 columns (for example).
Tanel Poder provides a nifty little tip on how to use a hidden parameter (_dump_qbc_tree) to resolve object and column names in SQL to the base tables and columns in Oracle.
If you are looking for some Oracle SQL coverage, an educational overview of the DECODE function is available in Bilal HATİPOĞLU’s blog this week. So if you are looking for help on using IF-THEN type logic from within an Oracle SQL script, check this blog post out because it contains a syntax diagram, sample code, demos, and even links and references to follow for more research on the topic.
And finally (for Oracle), Eddie Awad provides some clarification on the PL/Scope database corruption issue he blogged about last week.
SQL Server
Over at Database Underground Sean McCown extols the virtues of tiered storage and solid state disks for Microsoft SQL Server databases.Sean also posted a brief blog entry to remind SQL Server folks about this year’s PASS conference:, which is finishing up as I post this. It ran from September 18 thru 21, 2007 in Denver, CO, so if you are interested and missed out this year, it is not too late to start planning for next year!
The good folks at Thinking Outside of the Box have posted some SQL for finding records in one table not present in another table.
If you want some basic education on data clustering for SQL Server, check out Mladen Prajdic’s informative blog entry titled Back To Basics: What is a Clustered and a Non-Clustered index. He does a nice job of presenting a high-level description of clustering and its benefits.
MySQL
In MySQL news this week, The Data Charmer, Giuseppe Maxia, offers us the MySQL Test Creator -- a tool that will speed up and simplifythe writing of test scripts, during a normal session with the MySQL client.
Peter Zaitsev talks about the always-intriguing topic of buffer tuning in the MySQL Performance Blog. He highlights his benchmark results for various read_buffer_size values what read_buffer_size value is optimal? Later in the week Peter followed up with another excellent post on MySQL performance titled Using GROUP BY WITH ROLLUP for Reporting Performance Optimization.
And Lewis Cunningham is branching out. It would seem that the esteemed Oracle ACE is extending his blogging into the MySQL world. This week he discusses Partitioning in MySQL 5.1 in the MySQL Database News blog.
Additional Data-Related Blog News
And we had some interesting data and database blog posts this week that were not specific to a single DBMS, too. First we have Chen Shapira whose blog, I’m just a simple DBA on a complex production system, is always entertaining. This week she muses on an unconventional use for a database.
The Crazy DBA rants on this week wondering Where Have All the DBAs Gone? This amusing and thoughtful screed ruminates on the sorry state of database administration practices and procedures implemented in many of today’s applications.
James Koopman offers a handy treatise on handling NULLs in his blog, An Expert’s Guide to Database Solutions. If nulls give you trouble or cause confusion you should certainly check out James’ excellent overview of the topic.
Michael Hillenbrand at Chicken Soup for the DBA’s Soul offers up 10 Guiding Principles of an IT Manager based on his years of experience managing DBAs.
And finally, if you are involved in data governance project or are at all concerned with managing data as a valuable corporate asset, then be sure to keep an eye on The Data Governance blog. The Data Governance blog is written by a Governance Program Manager for a large financial organization and it covers the issues, trials, and tribulations of setting up and running a data governance program. An interesting entry that provides some helpful guidance on data governance and change management is titled 3 Ways to Catch Changes to Governed Data Before they Occur.
Thanks for taking the time to read this week’s Log Buffer. I learned quite a lot as I had to read a ton of blogs this week. Hopefully you learned a thing or two, as well. And be sure to tune in again next week, as Log Buffer #64 is being put together by Keith Murphy on his Diamond Notes blog.