MySQL

We ran two free, open sourced database benchmarks for this portion of the analysis.  The first uses MySQLd 4.0.20d installed from x86_64 RPM; precompiled by the SuSE team.  We ran the test-select and test-insert sql-bench exams.

MySQL 4.0.20d - Test-Select

MySQL 4.0.20d - Test-Insert

Postgres

Adding Postgres to the SQL benchmarks was fairly easy.  Although this may not be the most optimized way to benchmark Postgres, our time was limited.  We installed the DBD-pg Perl library, and then reran our sql-bench results as above in the MySQL tests Postgres was started with the "-o - F" flags. 

PostgreSQL 7.4.2 - Test-Select
PostgreSQL 7.4.2 - Test-Insert

Remember, even though times are significantly longer in this benchmark, they do not show whether or not one database is faster than the other.  However, there are some extreme conclusions we can draw from this test.  The Opteron 150 really advances over the Xeon processor in this benchmark.  We have shown in previous reviews that HyperThreading seems to hurt MySQL performance as some unnecessary threading probably occurs. 

Index Chess Benchmarks
Comments Locked

92 Comments

View All Comments

  • JGunther - Thursday, August 12, 2004 - link

    Yeah... this reivew (to me) proves that Kris is a good, well-intentioned guy, as he put aside his own personal time to re-do these benchmarks. But the results within also prove how utterly inaccurate the first review was, thus justifying (some of) the criticism he recieved.

    I can see that you did learn at least one lesson, Kris; there are no claims in the conclusion of the Opteron "trouncing" the Xeon this time (even though such a remark may be justified now). :)
  • thatsright - Thursday, August 12, 2004 - link

    Now will all of you A-Holes get off KrizK's & AT editorial staff's back!!

Log in

Don't have an account? Sign up now