This Thursday (September 24th, 14:00 UTC), Heikki Tuuri, the father of InnoDB, will give a session on Concurrency Control: How It Really Works. He'll describe how InnoDB manages concurrency control, so that the system protects data integrity. Beginning with the basics of transaction management, Heikki will include a discussion of the ACID (atomicity, consistency, isolation, and durability) properties, and explain various transaction modes, locking, deadlocks, and more advanced topics such as the impact of next-key (gap) locking, referential integrity, XA (distributed transaction management) support, and more. While the discussion will focus on the InnoDB implementation, many of the concepts presented apply to other database systems and storage engines.

For MySQL …

Last Thursday (September 17, 13:00 UTC), Lars Thalmann explained the Architecture of MySQL Backup. Lars is leading the MySQL Replication & Backup teams, and has given several MySQL University sessions before.

I was on sick leave last week and forgot to announce this session - sorry!  However, since the session was recorded (video & audio), you can listen to it anytime. Please find the recording and the presentation slides on this page.

MySQL University is a free educational online program for engineers/developers. MySQL University sessions are open to anyone, not just Sun employees. Sessions are recorded (slides and audio), so if you can't attend the live session you can look at the recording anytime after the session.

Here's the …

Mysqlslap is a good benchmarking tool but can be much more versatile in my opinion. So, recently, I was benchmarking a problem that was, not hard to solve, but somewhat tricky to benchmark. Using mysqlslap was the fastest way to get the answers I wanted with the least amount of BS, or so I thought!

Lets take a look:

-- The problem
1. A MySQL instance with 5 databases
2. Only one of the databases looked to be the problem from system analysis
3. Light concurrency issues suspected, given that the tables were MyISAM and the information gathered from the customer

-- How to run a close to real world benchmark
Turning on the general query log is a must in this situation so I gathered 1 hour worth of transactions. Please keep in mind that you have to take a snapshot of the data before you …

This Thursday (September 24th, 14:00 UTC), Heikki Tuuri, the father of InnoDB, will give a session on Concurrency Control: How It Really Works. He'll describe how InnoDB manages concurrency control, so that the system protects data integrity. Beginning with the basics of transaction management, Heikki will include a discussion of the ACID (atomicity, consistency, isolation, and durability) properties, and explain various transaction modes, locking, deadlocks, and more advanced topics such as the impact of next-key (gap) locking, referential integrity, XA (distributed transaction management) support, and more. While the discussion will focus on the InnoDB implementation, many of the concepts presented apply to other database systems and storage engines.

For MySQL …

This Thursday (September 24th, 14:00 UTC), Heikki Tuuri, the father of InnoDB, will give a session on Concurrency Control: How It Really Works. He'll describe how InnoDB manages concurrency control, so that the system protects data integrity. Beginning with the basics of transaction management, Heikki will include a discussion of the ACID (atomicity, consistency, isolation, and durability) properties, and explain various transaction modes, locking, deadlocks, and more advanced topics such as the impact of next-key (gap) locking, referential integrity, XA (distributed transaction management) support, and more. While the discussion will focus on the InnoDB implementation, many of the concepts presented apply to other database systems and storage engines.

For MySQL …

Last Thursday (September 17, 13:00 UTC), Lars Thalmann explained the Architecture of MySQL Backup. Lars is leading the MySQL Replication & Backup teams, and has given several MySQL University sessions before.

I was on sick leave last week and forgot to announce this session - sorry!  However, since the session was recorded (video & audio), you can listen to it anytime. Please find the recording and the presentation slides on this page.

MySQL University is a free educational online program for engineers/developers. MySQL University sessions are open to anyone, not just Sun employees. Sessions are recorded (slides and audio), so if you can't attend the live session you can look at the recording anytime after the session.

Here's the …

Last Thursday (September 17, 13:00 UTC), Lars Thalmann explained the Architecture of MySQL Backup. Lars is leading the MySQL Replication & Backup teams, and has given several MySQL University sessions before.

I was on sick leave last week and forgot to announce this session - sorry!  However, since the session was recorded (video & audio), you can listen to it anytime. Please find the recording and the presentation slides on this page.

MySQL University is a free educational online program for engineers/developers. MySQL University sessions are open to anyone, not just Sun employees. Sessions are recorded (slides and audio), so if you can't attend the live session you can look at the recording anytime after the session.

Here's the …

The query cache has had an interesting history with MySQL. I don't have much experience with it. I might want to use it in the future so I enabled it during a run of sysbench readonly on an 8-core server. The results are fascinating. Performance with the query cache enabled is bad for MySQL 5.0.44, good for MySQL 5.0.84 and then bad again for MySQL 5.1.38. By bad I mean that throughput does not increase with concurrency as it does when the query cache is disabled.

This lists the throughput in the number of transactions per second reported by sysbench readonly using an 8-core server. It was run for 1, 2, 4, 8, 16 and 32 concurrent sessions.

Throughput with query cache enabled:
            1     2     4     8    16    32
5.0.44    663   746   872   834   730   694
5.0.84    639  1033  1819  3216  2636  2466
5.1.38    662   725   818   782   731   699

Throughput with query cache disabled:
            1     2     4     8    16    32 …

Mike Hogan, CEO of ScaleDB spoke at the Boston MySQL User Group in September 2009:

ScaleDB is a storage engine for MySQL that delivers shared-disk clustering. It has been described as the Oracle RAC of MySQL. Using ScaleDB, you can scale your cluster by simply adding nodes, without partitioning your data. Each node has full read/write capability, eliminating the need for slaves, while delivering cluster-level load balancing. ScaleDB is looking for additional beta testers, there is a sign up at http://www.scaledb.com.

Slides are online (and downloadable) at http://www.slideshare.net/Sheeri/scale-db-preso-for-boston-my-sql-meetup-92009

Watch the video online at …

Interviewed at an industry gathering in Silicon Valley, Oracle CEO maintains that despite EU concerns, Oracle's database does not compete with MySQL.