Take a look here:

  Response Time (s)
 Transaction      %    Average :    90th %        Total        Rollbacks      %
------------  -----  ---------------------  -----------  ---------------  -----
    Delivery   3.98      0.211 :     0.266       274829                0   0.00
   New Order  44.78      0.157 :     0.187      3090951            30925   1.00
Order Status   3.99      0.149 :     0.179       275357                0   0.00
     Payment  42.76      0.150 :     0.180      2951361                0   0.00
 Stock Level   3.99      0.152 :     0.182       275564            92070  33.41

50606.82 new-order transactions per minute (NOTPM)
60.5 minute duration
0 total unknown errors
31 second(s) ramping up

If you know what this output is from, and you know what 50K TPM means… your probably curious about these #’s. I am probably tantalizing you right now in fact. But I am not going to tell you more, not yet. So go ahead and guess. Better yet come …

This is an issue that keeps rearing its ugly head over and over again, and since it greatly affects performance, it is most important that DBAs of any DMBS running on Linux come to grips with it. So I decided to do some research and try different settings on my notebook. Here are my findings.

What can you find on the web?

A Wikipedia search for the word swappiness will come up empty (any volunteers out there want to write an article?). A Google search will show some pretty old material—the best article I found is from 2004: Linux: Tuning Swappiness. This article includes a detailed discussion with some interesting remarks by Andrew Morton, a Linux kernel maintainer.

So, what is swappiness?

Towards the end of the email thread quoted in the article, you’ll find this definition (sort of):

> I’ve read the source for where …

Only a little over a week before the User conference and I am still burning the midnight oil to get as much information for my presentations as possible. I thought I would tease you a bit here. What do you get when you put 4 Intel X-25E’s ( Sun branded) SSD’s running RAID10 in a Sun 4450 and run the sysbench fileio test on it?

We have finally uploaded the next beta packages of our upcoming database design tool. We have fixed several bugs and added some enhancements that to the Interface. There’s a new editor for defining/maintaining User Defined Datatypes.

There are binary-packages for Windows, Mac OS X Leopard and Linux (Ubuntu and Fedora) as well as the zipped sources for building Workbench yourself. If you experience problems please let us know by filing bugs at http://bugs.mysql.com or contact us directly on via irc-client on irc.freenode.net channel #workbench.
Thank you all for your feedback and contribution. Fetch the new packages at our main download page and give it another try.

A little while ago, I was the brave soul tasked with making sure Drizzle was working properly and passing all tests on Solaris and OpenSolaris. Brian recently blogged about some of the advantages of also running on Solaris and the SunStudio compilers - more warnings from the compiler is a good thing. Many kudos goes to Monty Taylor for being the brave soul who fixed most of the compiler warnings (and for us, warnings=errors - so we have to fix them) for the SunStudio compilers before I got to making te tests work.

So, I got to the end of it all and got pointed to an OpenSolaris x86 box where the drizzleslap test was timing out. The timeout for tests is some amazingly long amount of time - 15 minutes. All the drizzle-test-run tests are rather short tests.

To make running the tests quick, I usually LD_PRELOAD …

Some weeks ago Luke Kaines stated his observation that “requiring copyright attribution is a greater sin than providing commercial add-ons”.

His perspective was based on the theory that requiring copyright assignment restricts the developer community, a theory that was apparently repeated by Dave Neary during the recent OSBC event (I missed that session due to our CAOS client lunch).

Daniel Chalef of KnowledgeTree provided some evidence of Dave’s perspective and also the contrary view that the assignment of copyright is critical not just for vendor-dominated projects but also for community projects such as The Mozilla Foundation, The …

Continuing on from CPU, we turn our attention to Memory. One of the first steps when addressing a MySQL performance tuning problem is to perform a system audit of the physical hardware resources, then identify any obvious bottlenecks in these resources.

In auditing, I start with the ‘free’ command, the already used ‘vmstat’ command, the /proc/meminfo and /proc/sys/vm/swappiness files to get an indication of memory and swap resources. While we are looking at the Memory, the configuration of Swap is also very important. I will discuss this in more detail later.

$ free -m
             total       used       free     shared    buffers     cached
Mem:          3955       3838        117          0        402       2366
-/+ buffers/cache:       1069       2886
Swap:         1027          0 …

On early Tuesday morning, I made a quick trip to Paris, France, to attend and speak at the Solutions Linux / Open Source 2009 Conference. I've never been to this conference before and was quite surprised about its size - it's actually the largest Open Source event in France and it reminded me a lot of LinuxTag in Germany. Many well-known vendors (e.g. Sun, Novell, Canonical, Bull, etc.) were exhibiting. The also was a large "DotOrg" section for various Open Source projects and I was very happy to see that LeMUG.fr, the official MySQL User Group of France, had a table there, too! A big Thank You goes to Pascal Borghino, who manned that table on his own most of the time and answered questions about MySQL. I …

Assuming you have successfully compiled Drizzle, and you are ready to start for the first time, here are some beginner differences with those familiar with the current MySQL 5.1 GA version. Note: drizzle is actually an original fork of the 6.0.x codebase however the code looks very little like mysql anymore.

Getting started differences

The first default MySQL installation step scripts/install_mysql_db is gone, and there is no replacement. The ‘mysql’ schema no longer exists. Talking about schemas, ‘test’ is also gone by default.

While a bin/drizzled_safe exists to start drizzle, this presently doesn’t operate as expected. You should start drizzle with sbin/drizzled &

They still have an empty …

Taking the work already done with vmplot.sh, a useful tool for MySQL performance tuning by Yves and Matt at BigDBAHead, and in true Open Source fashion I’ve enhanced and modified for my own purposes.

These changes include:

• Error checking for ‘gnuplot’ command on the system
• Eliminate the first row of sample data, as this is often not a complete sample for the vmstat duration.
• Created a HTML output file for easy browser viewing
• Changed Memory scale values from Kilobytes to Megabytes
• Resizing png’s for optimal 1024×768 display output (2 per row)

Download vmplot

And I get:

So more specifics of what I did.

Install gnuplot.

$ yum install gnuplot

Create vmstat sample …