OT: Things Fashion Stores and Designers just don’t Get

I have just returned from a trip to Milan, my lovely wife traveling with me. And after a full weekend there, I feel like ranting again. For my trusted readers, if you expect something technical from this, please stop reading now.

Curious Partition Function Behaviour

Just another short blog today describing a curious issue I found with a query plan this week and a “workaround”.

In our core system, we have a table with two partitions. One partition contains all the work that “has been done” (which has the column WorkItem set to –1) and the other the “work   in progress” (with WorkItem to different values, all > -1).

The reason we have created just two partitions for this table (which is a heap) is that the items that are “work in progress” are often scanned, yet the work that has been done (WorkItem = –1) is the vast majority of the table. This “mini partitioning” is a nice design pattern I often apply to skewed distribution like these. It provides a significant performance boost on table scans. But this week I saw an oddity I have not run into before.

More Courses Available – in Sweden and UK!

I am happy to announce that my tuning course and newly developed Data Warehousing course is now available.

Tuning/Scaling Course

The tuning course will be hosted by CrossJoin Consulting in London, UK You can sign up for it here:

• Tuning Course in London (13th June 2013)

You can still make it for the early bird rates. This course is probably THE deepest level course about SQL Server out there. You will learn the nitty-gritty details of xperf profiling, spinlock detection and multi-threaded optimisation.

We are going to be digging deep below the surface of SQL Server here, be prepared for an intensive day.

 

Data Warehousing Course

The next data warehousing course will be in Kista, Sweden. It will be hosted by SolidQ and you can sign up for it here:

• Data Warehousing in Sweden (7th February 2013)

This will be a very unique course with both Davide Mauri and myself teaching. Its a one day intensive training in my usual “no nonsense” style where you learn about data modeling from two of the leading experts in the field. The course extends the ideas I have blogged extensively about here.

There are still seats left, but they are going fast.

I look forward to seeing you there!

First publicly available courses –in Aarhus, Denmark – 23-24 October

I am pleased to announce that both my Tuning Course and the Data Warehouse Modeling course will be run in Aarhus Denmark. The courses will be held in English, even though they are hosted in the very city I was born.

The course will be hosted by Orangeman, who will be handling the logistics. There are still available spots and if you act fast, you can get the early bird price.

The relevant details:

• 23rd October 2012: Tuning Course
• 24th October 2012: Data Warehouse modeling

See you there for some intense days of tuning and modeling.

Open for Business!

Today, 22nd June, is my last day at Microsoft. This means that I will be open for business from Monday 25th June 2012.

I will take on short term consulting contracts (including subcontracting if you need it). The knowledge I offer extends all the way from hardware and OS scaling and tuning, to architecture guidance, data modeling and ongoing monitoring and coaching during your implementation process to help build a technically excellent solution. Details of my skills are available here: CV.

In addition to consulting, I will offer a one day course on high scale performance tuning. Course details will follow on this blog next week. I can already now reveal that the course will feature xperf tuning, data modeling tricks, tuning methods and  a deep understanding of how to squeeze out more performance from modern multi-core machines.

If you are interested in purchasing my services, please send enquiries to: thomas@kejser.org.

Best wishes
Thomas

My official blog URL

You can now access my blog at http://blog.kejser.org. Please point permalinks to that domain.

Resurrected!

I have decided to resurrect my old blog. There are new battles to fight and information to share.

First of all, a summary of what I have been doing the last 4 years:

I joined the SQLCAT team working on the largest SQL server installations I the world. It is exciting stuff and as far as I know, I have worked on:

• The largest single SQL database in the world: at 75TB
• One of the fastest core banking systems around: 13K business tx/sec, mapping to 52K database writes/sec (2 inserts, 2 updates per tx)
• The most scaled cube in the world: 1000 concurrently executing MDX queries in a 4TB cube racking up 50K IOPS per server – 4 servers
• The biggest SQL scale out system around – at 1PB using DPV
• Tuning for the ETL World Record
• Some of the fastest I/O systems ever attached to SQL server

I have tried to keep up with publishing whitepapers as I went along, sharing all I could. Two notable papers come to mind:

• The Data Loading Performance Guide
• The Analysis Services 2008 Performance Guide

Along with this, a ton of information on SQLCAT.com in the form of blogs, top 10 lists and TechNotes.

I wanted to bring back this blog, because I think there are a lot of things worth talking about that can be said without going through the somewhat long publishing process of Microsoft. For example, one of my big passions these days is good data modeling, without which the above could never have been achieved. I am almost sure that some of the discussions I will start on this subject will step on some feet (Data Vault modelers, consider yourself warned)

So here I am, back again. And let me stress right away: the opinions expressed on this blog are my own, not those of Microsoft.

Location:Helsinki, Finland

Moved to WordPress

I have moved my old blog to WordPress. I still dont expect to be active here, but at least the old information is now preserved in cyberspace.

Comments have been disabled to avoid spamming. (Thanks Anne)