The Wayback Machine - https://web.archive.org/web/20090307032257/http://blogs.sun.com:80/marchamilton/entry/how_much_are_you_paying

Marc Hamilton's Weblog

How Much Are You Paying For Your Database?

Thursday Dec 13, 2007

Allen Packer's recent blog titled, Are Proprietary Databases Doomed?, talks about the new challenges to traditional proprietary commercial databases by open source upstarts like MySQL and PostgreSQL. I'd like to point out that every free copy of Solaris 10 ships with optimized versions of both PostgreSQL and MySQL, along with JavaDB. Still, while I agree with many of the points in Allen's blog, there are plenty of commercial applications (many owned by the database vendor) that have, in effect, hardcoded ties to a commercial database, so I'm not too worried about those commercial databases being doomed. Sun spends a tremendous amount of effort working with both Oracle and IBM DB2 to ensure they run great on the Solaris platform, both on SPARC and x86.

Yet if you are not locked into a commercial database, it is hard to ignore the world record performance of PostgreSQL running on Solaris. Maybe that is why it is hard to find a startup these days that is writing big checks to Oracle. If you want commercial support for your database, Sun's PostgreSQL support starts at only $1596/year. Why should you turn to Sun for your PostgreSQL support? Well for one, we use it ourselves in many Sun products that require a high performance database, like Sun xVM, so you can be sure we work hard to optimize and support it.

We are also seeing PostgreSQL turn up in more and more ISV solutions. Greenplum for instance, uses PostgreSQL as the basis for their revolutionary data warehouse appliance (which happens to be based on Solaris and ZFS). One recent customer which evaluated Greenplum versus traditional commercial data warehouse solutions found that the energy cost savings alone of a Greenplum solution running on Sun's Thumper storage server paid for the cost of the software in the first year! So maybe the question should be not how much do you pay for your database, but how much power can you save with a more efficient database?

Like this post? del.icio.us | furl | slashdot | technorati | digg
Comments:

Post a Comment:
Comments are closed for this entry.