Recently in PostgreSQL Category

Unless you've been under a rock, you know Sun is buying MySQL AB. That was news, last week. But it was also rather puzzling news, in many ways, and worth taking some time out to consider the implications.

So, for Part One of my coverage of the Sun/MySQL deal, I'm mostly going to offer a bunch of links. For example, the Yahoo! News breaking news story: Sun to buy MySQL for $1 billion. You get the basics of the deal here, but you also get some of the issues spelled out for you: Sun is a "server maker"; meaning, they're a hardware company. Other key points to consider:

• Sun expects to spread MySQL into big corporations, who already buy Sun hardware, as well as bolster existing Sun partnerships such as those with IBM and Dell.
• Sun is a software company, too, and an open source software company at that. They, like IBM, see offering open source software as a selling point for their hardware.
• MySQL is dwarfed by the competition from Oracle and Microsoft who dominate in the traditional database market, but MySQL owns the web database market, which is growing (while the more traditional market may be stagnating).

The Yahoo! article gives a pretty good precis of the situation, but you need to dig a little deeper to figure out what exactly is going on. For example, they don't say much about the extent to which Sun controls some very important pieces of software already, including Java, OpenSolaris, StarOffice, NetBeans, and more. Nor do they mention Sun's existing relationship with the other star open source DBMS project, PostgreSQL. For more about Sun, MySQL and PostgreSQL, check out Sun buys MySQL - what about PostgreSQL?, and Sun Purchasing MySQL and PostgreSQL advances from PostgreSQL.

Check back next week for my take on the implications and impact this deal with ultimately have, but in the meantime, here are some of the more interesting articles/blog entries I've found in the past few days about it:

• Ars Technica has a "breaking" story w/ some commentary: Sun targets open source LAMP stack with MySQL acquisition
• Here's the MySQL blog entry announcing the acquisition: Sun acquires MySQL
• Betsy Schiffman at Wired seems to think the deal is overpriced, in Sun Pays Up the Wazoo -- $1 Billion -- for MySQL; she cites analyst estimates of MySQL revenues being just $75-80 million, rather low for a $1 billion deal.
• Dvorak likewise doesn't mince his words; in The Sun-MySQL deal stinks he suggests the deal is horrible (for MySQL) and the only winner will be Oracle (!).
• More from MySQL blogs here Talking to MySQL Founders Monty and David on Sun
• MySQL and Sun - my thoughts from David Van Couvering

For the record, the official press releases here:

• Sun's press release, Sun Microsystems Announces Agreement to Acquire MySQL, Developer of the World's Most Popular Open Source Database
• MySQL's press release: Sun to Acquire MySQL

Benchmark testing can be expensive, and proprietary software licenses often restrict what kind of benchmark testing you can do, and how/whether you can even publish your results. So it's nice to see this blog item from Josh Berkus about PostgreSQL benchmark results.

In case you didn't know, Josh Berkus is a core member of the PostgreSQL development team.

About that benchmark: here's a link to the published benchmark report. According to Josh, the key take-aways from this report are:

• This is the first real (that is, formal) PostgreSQL benchmark to be published. Josh notes that:
  

  
  only an industry-standard and peer-validated benchmark is sufficient to assure potential users that they're making the right choice of platform. Given some of my conversations with large corporate customers, I expect this benchmark to influence corporate software buyers.



• The results should help disprove the misconception that PostgreSQL is sluggish compared to MySQL or Oracle.


• The results should also demonstrate the degree to which PostgreSQL running on commodity hardware can be very, very competitive with Oracle running on more expensive brand-name machinery.

Just a quick update/roundup on the latest PostgreSQL action:

• In case you were wondering about real-world applications that use PostgreSQL, check out their PostgreSQL case studies links to narratives of how real companies are using PostgreSQL.
• Want more examples? Check out the list of Featured Users of PostgreSQL.
• Need help? Read this article on Performance Tuning PostgreSQL by Frank Wiles at Revolution Systems, an open source consultancy.
• More help is available from Varlena, LLC, another open source database consultancy: Tuning PostgreSQL for performance. This one's a little old, but still worth reviewing.
• Here's another performance tuning checklist, PostgreSQL 8.0 Performance Checklist, from Power PostgreSQL, the website for a book by Josh Berkus and Joe Conway, well-known PostgreSQL experts.

Let me know if you've run across any interesting PostgreSQL stories or links.

Jonathan Ellis ran an interesting note about MySQL and PostgreSQL database benchmarking (Benchmark: PostgreSQL beats the stuffing out of MySQL), with pointers to some work done at the Netherlands tech website, Tweakers.net.

Jonathan points out that while MySQL's own benchmarks tend to favor MySQL (just like any vendor and their own internal benchmarks), he points out that when the benchmarking is being done by independent third-party, PostgreSQL seems to do significantly better.

Benchmarking is an interesting field, producing interesting results, but performance on standardized tests must be balanced against the performance on the actual task in question as well as the suitability of the product to meet that particular challenge.

The original content (in English, thankfully), is here and here.

Whether you're considering PostgreSQL as an alternative SQL solution or want someone higher up in the command chain to consider it, this article could help: Five reasons why you should never use PostgreSQL -- ever.

Warning: ignore that title. The article is really about five of the top reasons that people think PostgreSQL isn't going to work for them, and how to answer those objections. The article's author, W. Jason Gilmore, handily dispels the fear, uncertainty and doubt associated with using PostgreSQL rather than some commercial product, but his answers to objections such as "it doesn't run on Windows", "there's no professional tools", and "there is no support" can just as easily be applied to any number of other open source software projects.