It's the second point release of MySQL Enterprise Monitor 2.3. And there some nice new features and bugfixes.

It's now possible to enable or disable event blackout for specific servers easily.

It should now graph InnoDB with multiple buffer pools correctly. It won't work for 5.5.8 however, see Bug #60777 and Bug #60831 for more info on that.

Read the complete changelogs for all the details.

The Advisors/Rules were also updated. But there is still room for improvement:

• Bug #60828 Add rule to detect duplicate foreign keys
• …

I’m giving two talks tomorrow (Tuesday) at the MySQL Conference and Expo:

HailDB: A NoSQL API direct to InnoDB, 2:00pm, Ballroom D

Dropping ACID: Eating Data In A Web 2.0 Cloud World 3:05pm, Ballroom G

The HailDB talk is all about a C API to embed an InnoDB based relational database engine into your application. Awesome stuff (also nice and technical).

The second talk, “Dropping ACID: Eating Data in a Web 2.0 Cloud World” is not only a joke that only database people get, but a humorous and serious look at data integrity and reliability as promised by the current hype. This was quite well received at linux.conf.au in January. So, if you weren’t in Australia in January this year, then certainly come along and see how you go …

I had a quick look at the source tree (I haven’t compiled it, just read the source – that’s what I do. I challenge any C/C++ compiler to keep up with my brain!) that’s got a tarball up on labs.mysql.com for the memcached interface to innodb. A few quick thoughts:

• Where’s the Bazaar tree on launchpad? I hate pulling tarballs, following the dev tree is much more interesting from a tech perspective (especially for early development releases). I note that the NDB memcached stuff is up on launchpad now, so yay there. I would love it if the InnoDB team in general was much more open with development, especially with having source trees up on launchpad.
• It embeds a copy of the memcached server engines branch into the MySQL tree. This is probably the correct way to go. There is no real sense in re-implementing the protocol and network stack (this is about half what memcached is anyway).
• The copy of the …

I just attended the Facebook Open Compute Project tech talk and I was very impressed. I am located in Santa Clara for this week whilst attending #mysqlconf and thought to swing by Facebook and attend this talk, partly because I wanted to find out more about their server infrastructure.

The Open Compute Project is driven by Facebook as a way to innovate in datacenter excellence for all.

Tom Furlong, Director of Site Operations and Jay Park Head of Datacenter Design highlighted the Facebook 2.0 DC which includes

• fully custom servers
• fully custom cooling for the datacenter
• completely new power distribution within the datacenter
  

Facebook have multiple DC's around the USA and use Akamai to serve content locally …

The Need for Multi-Threaded Slaves! The MySQL replication team has come up with a feature that looks very promising for improving slave scalability. Although there is some work that one can do on the master side, it is the work already done at the slave side that this blog post highlights...

At its core, MySQL replication is single-threaded! In detail, the tiniest unit of work for replication is an event, and a group of events forms a transaction. Events are pushed by the master to the slaves by a thread, known as "dump thread". At the slave, a reader ("IO thread") reads event-by-event and writes them to a local persistent queue, the "relay log". Then a single threaded applier, the "SQL thread", reads and applies events sequentially.

Contrary to the master, which executes transactions concurrently, the slave serializes execution of each and every transaction, even if they were executed concurrently on the master and are …

 Just got done with day 1 from the MySQL conf 2011... tutorial day. I decided to attend a session a bit outside of my comfort zone, so I chose MySQL Plugin Development. I haven't written any serious C/C++ for over 10 years, but lately it's been more appealing to me.

read more

This is the third part of the MySQL 5.6 Performance Notes started from here.

Now let's get more in depth and analyze performance results on more "focused" workload scenarios.. - for the moment MySQL/InnoDB is scaling up on 32 cores and most optimally working with up to 32 concurrent sessions where no any severe contention is not yet happening. So the 32 users workload on 32 cores will enter in the focus in the following part, but this time it's a full troubleshooting rather benchmarking.. ;-))

NOTE: there was a question what is the difference between "Read+UPDATE" vs "Read+Write" test? - "Read+Write" is doing 3 DML queries within a one operation (in reality there are 3 transactions within 1 Write) - DELETE, INSERT and UPDATE. While "Read+UPDATE" is executing only UPDATE query within Write operation, …

Hello to those of you who are in Santa Clara!  Patrick Galbraith and I will be here this week, speaking and glad handing.  You can find us both in the Attendee Directory and identify either or both of us as someone you’d like to meet.

Patrick’s Sessions are:

5:15pm Wednesday – Perl programming with MySQL and Drizzle
11:55am Thursday – Narada: using Gearman, Sphinx, Memcached, Drizzle and PHP to build a search application

I’ll be speaking in Wednesday night’s Ignite MySQLconf and at:

10:50am Thursday – …

Without a doubt Debian versions of Linux are becoming more and more popular. (Ubunto and Suse for example)  It is not rocket science to get a MySQL server installed on a Debian system.
 apt-get install mysql-server
The problem some people have is making it easy to get the latest version of MySQL and not just the version that is available via the Advanced Packaging Tool (apt).
 I wanted to at least make available a script/steps for everyone to get the job done quickly and easily.  I do not take all the credit for this. I did find a lot of the work done here , I just packaged it up. 

Cut and paste the following into a script.  ( mysql_source_install.sh )  chmod +x the script ( chmod +x mysql_source_install.sh )  and then execute it …

We ran a number of tests to see how many reads per second we could get from MySQL Cluster. We used a modified version of flexAsynch (as shown in previous blog), where each record read was 100 bytes in size.

With a cluster of 4 data nodes operating on 2 machines we were able to process 1.15M reads per second. On a cluster consisting of 8 data nodes executing on 4 machines we were able to process 2.13M reads per second. On a 16-data node cluster with 8 machines used for data nodes, we were able to process 4.33M reads per second and finally a cluster with 32 data nodes distributed on 16 machines we executed 6.82M reads per second. The tests were run on MySQL Cluster 7.1, we're confident that similar numbers can be achieved with MySQL Cluster 7.0 and also with the new beta version MySQL Cluster 7.2.

This benchmark will give you a good idea what can be achieved with direct usage of the NDB API, and using other APIs like …