Update: Has SQL Server already lost Mind Share?

In the comments on my previous post, Erdju made the suggestion that I add StackOverflow to my analysis – suggesting that the more SQL Server questions might be found there, changing the result of the analysis. Using the data.stackexchange.com site to analyse data as suggested by Nick Craver (instead of my own, mocked up version) I can now extract recent data for both sites.

This posts contains the updated result as well as a few additions to the previous post.

Continue reading…

is SQL Server Losing Mindshare?

As many of you have noticed, I have been flirting a lot with open source database lately. I am no longer spending much time going to SQL Server conferences.

About two years ago, I decided that it is time to diversify my knowledge and spend more time with other database products. Back then, this was largely based on a hunch that Open Source is now getting to the point where it is worth looking into. I didn’t really have any data to back up my decision to lower my investment in SQL Server – it was mostly intuition, a feeling for the zeitgeist.

Today, I wanted to assess where things are. It is difficult to get information about the popularity of databases that is not heavily biased, so I decided to mash up some data myself.

Continue reading…

SQL Server Logo

Database Page Row Order

Does Physically ordering rows inside database pages make sense?

A few weeks ago I was asked a great question by Sam, who follows my blog. With Sam’s permission, I am reprinting the question here.

Sam asks (my highlights) :

We are told that we can have only one Clustered Index since we can actually sort the data in only one way. However, a clustered index orders data in the leaf pages in sequential order, but does not order the data within that page itself to be in sequential order. Since the data on the page itself is not sorted sequentially (and thus implying more than one way of possible order), I am confused by the “Since we can sort data only in one way we can’t have more than 1 clustered index” reasoning.

Continue reading…

Prefer Mobile Data over WIFI on Mac OSX

With both 3G and 4G becoming more prevalent, I often find myself getting a better connection over an my mobile data subscription than via a public hot spot.

The problem you often get is that an unstable WIFI hotspot will make your Mac appear unresponsive, even if you have a perfect mobile data connection on a USB or bluetooth connected iPhone.

In this blog, I will describe a simple trick that will configure your Mac OSX to prefer a USB connected iPhone over a WIFI hotspot when both are available.

Continue reading…

Domino pieces about to fall

Adding another transaction log file to Gain Performance

Mike West recently posted a rather active thread on LinkedIn about databases with more than one transaction log file. Eventually, I realised that the full answer to the question: “When does it make sense to add another log file to a database for performance reasons?” is complex enough that it needs a blog available for future reference.

Continue reading…


TPC-H: Schema and Indexes

The TPC-H benchmark is often used a method for customers to evaluate data warehouse products to make purchasing decisions. Because it is such a crucial benchmark, it is important to understand the challenges it presents for database vendors. Unfortunately, the public information about tuning for TPC-H is rather sparse and it is generally hard to come by good documentation. Vendors do not like to be compared with other vendors – so their secrecy is understandable.

In this blog series, I will try to shed some light on the TPC-H benchmark, what I think is wrong with it, and provide some of my thoughts about the challenges you face when tuning it.

Continue reading…

A pen and handwriting

Make OSX and Linux command line behave more like Windows

Lately, I have been doing more and more work in my native OSX image. And I am really loving it.

However, being an old Windows user, I find a lot of habits hard to break and I constantly find myself wanting the standard Bash command line to behave just a little more like good old DOS. To avoid the same typing error over and over again, I found these modifications to be useful. I am sharing them here in hope they may help other people who are transitioning from Windows. Continue reading…

SQL Server Logo

On using Query Hints

The revealed wisdom in the SQL Server community has generally been that query hints should only be used as a last resort. For a complete novice of SQL Server, I would agree. But to call avoiding query hints a “best practise” is taking it too far.

For anyone with a little experience who knows what they are doing, I find that query hints are not only a good reactive solution, they are a proactive, design time, tool.

Continue reading…