Some news about a change in my career path that I've been meaning to announce. I now have moved on from Lycos and now work for NorthScale Inc!

In parting Lycos, I'd like to thank them for the great challenges I had while there. I designed and developed, along with my team members in the web publishing and support from the OPs group, PHP offerings for Tripod users, a long-awaited feature that premium users can use to install numerous PHP applications. Also, with that was a great interface I developed for installing applications. It required some work on the applications themselves to make them as easy to install as possible (similar to APS). The other task while there was to switch Tripod and Angelfire blogs from Oracle to MySQL. Thank you Lycos -- It was a pleasure working with all of you!

Now, I'm excited to announce I'm working with NorthScale. This is a great team of people -- including Dustin Sallings, Steve Yen, …

Asynchronous Replication for High Availability

MySQL replication is often positioned as a solution to provide extra throughput for your database (especially when dealing with high numbers of read operations). What tends to be overlooked is how it can be used to provide high availability (HA) – no matter how much redundancy you have at a local level, your system remains at risk from a single catastrophic failure – loss of power, earthquake, terrorist attack etc. By using MySQL asynchronous replication to maintain a geographically remote copy of that data, service can be maintained through any single site failure.

As this replication is asynchronous, there are no low latency demands on the network and so the two sites can be thousands of miles apart while utilising low-cost connections.

This article …

Departmental or subject-specific data warehouses – known as “data marts” in the industry – seem to be gaining in popularity.  Fueled partly by companies wanting to start small with focused projects in today’s economy, and partly by advances in data warehousing technology improving affordability and deployability, data marts seem to be popping-up everywhere.

In most cases, data mart projects are driven by the head of a business unit or a functional group (like Sales) needing to analyze their own slice of data in order to run their department more efficiently and effectively.  The data may come directly from an operational system or a combination of source systems resulting in what’s called an “independent data mart”, or it may come directly from a larger, enterprise data warehouse in a hub-and-spoke or “dependent data mart” configuration.

In either case today, according to industry analysts, companies …

270 folks took MySQL Certification exams in July and 118 earned certifications. This is our best month outside of our annual Users Conference. For the fiscal year, MySQL Certifications were up 150.27% over the previous year.

Since the program inception, we had almost 8,000 exams taken and over 3,300 certifications. Not bad but it pales in comparison to Java exams and certifications.

Prometric is our new testing partner. Those of you with unexpired Pearson VUE vouchers can request a exchange by contacting Certification@sun.com with your Voucher ID number. And those of you wishing to sign up for the exams (they are part of the Sun 310-series at http://www.prometric.com/Sun/default.htm) need to contact you local Sun sales office to purchase a voucher.

And the North Texas MySQL Users Group meets tonight -- see http://www.northtexasmysql.org for details.

Mat Keep and I will be presenting a Webinar on Wednesday, August 12, 2009 describing the new capabilities of MySQL Cluster 7.0. The time is designed to be Asia/Australasia-friendly but of course it’s open to everyone.

Register for the MySQL Cluster 7 Webinar here.

By attending this webinar, you will learn more about the next generation of MySQL Cluster, and how it enables you to deliver 99.999% database availability, with real time performance and linear scalability, while slashing TCO.

Boasting a range of breakthrough capabilities, MySQL Cluster 7.0 will enable you to stay ahead of your most demanding, mission-critical application requirements. Enhanced scalability delivers higher database performance with fewer nodes to simplify deployment and administration. Expanded platform support and interoperability delivers more …

Departmental or subject-specific data warehouses - known as "data marts" in the industry - seem to be gaining in popularity.  Fueled partly by companies wanting to start small with focused projects in today's economy, and partly by advances in data warehousing technology improving affordability and deployability, data marts seem to be popping-up everywhere.

In most cases, data mart projects are driven by the head of a business unit or a functional group (like Sales) needing to analyze their own slice of data in order to run their department more efficiently and effectively.  The data may come directly from an operational system or a combination of source systems resulting in what's called an "independent data mart", or it may come directly from a larger, enterprise data warehouse in a hub-and-spoke or "dependent data mart" configuration.

In either case today, according to industry analysts, companies are looking for data …

Last week I received a really extensive questionaire from the European commission. They investigate in the case of the Oracle/Sun merger and supplied a long questionaire how Mayflower sees the Oracle/Sun merger. One of the question, for example, is if the quality of InnoDB has increased or declined since Oracle bought Innobase Oy some years ago. I wrote to the commission and asked if I could distribute the two Word documents they delivered to me. Important information: the questionaire has to be sent back until August, 13th!

This is what Vera Pozzato on behalf of the team told me:

Dear Mr. Schotte,

Many thanks for your email and for the interesting information you have provided us with.

As our investigation is a rather extensive and complex exercise, we would appreciate if you could send us, if possible, the contact …

At a customer I have had the opportunity to work with some really state of the art equipment.

The customer is in the online gaming industry and we are building a platform based on MySQL Cluster where we are consolidating a number of applications. Workload is mainly inserts, updates, pk reads, and index scans (quite frequent), and a few joins. The cluster handles the transient data (current working set), and is replicating to a long term data store (following this approach), and we are also feeding other back-end support systems from this.

The cluster (>8 computers) is interconnected with Dolphin DX adapters. Computers are SUNs Nehalem …

There are many wrong ways to find a "random" row. For example:

SELECT .. ORDER BY rand() LIMIT 1;


The rand() function is great, and LIMIT is phenomenal, but the above query is really, really bad. It evaluates the rand() function for every row of the table and then (if that wasn't bad enough) it orders all the rows by the result of the rand function. Then it returns the first row. Wow. Don't make MySQL do that much work!

There are better ways to find a "random" row. Maybe in a future blog post I'll go into them, but at the very least consider calculating a random number ahead of time and looking up a row with that id.