The New Blue in Town…


It seems to be the season, where a lot of people switches jobs. At least in my social network LinkedIn states that 14 people have a new position, just this past month. Of course there are always those who head in a new direction every 24 months. And I’ve never really understood how they manage to establish themselves so quickly, but I seem to have a need to settle more in, when I finally decide on something. I held my previous position for 4 yrs, and my current position I have held for about 5½ yrs, until now.

Looking back, over my shoulder

During my years in Rehfeld, I have been engaged in several roles. I started out as a Business Intelligence developer doing custom dashboards for our clients. The customization was almost epic in proportion, in a way where the consultant in the end was no longer needed. Everything could be controlled from the data imported into the database (there were several databases, but only one (1) containing meta data) No, not that kind of meta data! But boy did I get smarter as I got along…
The bold directors then made an organizational change, in where I was to team up with a database developer and a .Net developer. The three of us were to distill the lessons learned in the organization during the last many years and make it into a product: Effektor. We wanted to build a toolbox, to enable the consultant to have the tedious parts of developing a DWH done by a click of a button, and at the same time with a very high level of quality. This across all major projects, so now every consultant would be able babysit others work, without the need of having to read up on how this dude implemented the T2 on that column, and why on Earth that implementation wasn’t the same as on these other two columns…
The consultant was the first job function the product should support. As of the latest version, we now support operations for security administrators, data stewards, data managers, super users and of course regular users.

Can’t Explain

In the last year+, I have been back into the field, as a Business Intelligence consultant, helping several clients on projects involving Data Mining, ETL, Server Upg., Custom Applications. I have been fortunate enough to have Effektor create much of the plumbing on several projects, and it really has made the development process a whole lot faster. As long as the business requirements are not too quirky or far out, the tool actually does most by meta data definition only. Thumbs up to Thomas Meltofte Eriksen (LinkedIn) and Mikael Nygaard (LinkedIn|Twitter) for doing such a smashing job on the tool. I really enjoyed our time on the PRO team.
As I left PRO, I joined Team SPOK (abb. for Stat, Privat og Kommune ~ Government, Private and Municipal). The team focused on clients in those sectors. There are a lot of team members that really do a great job everyday, are open-minded, happy and always willing to help and make everybody become a better developer. Some stand out, as I have worked with them on several assignments or do community work with them. Rasmus Reinholdt Nielsen (LinkedIn|Twitter),) whom I will be presenting at Campus Days with is a great co-worker/team lead whose wisdom and humor I am really going to miss. Luckily we get to meet, hopefully often, in the community. Rolf Veinø Sørensen (LinkedIn) who always has the demo of the newest stuff already done, sometimes even before the news hit your inbox. Finally Brian Bønk Rueløkke (blog|twitter) who would be a fantastic resource on any team. This guy really knows his stuff!
Hopefully I will be able to work with you guys in the future, sometime again. There’s just this something I need to do first…

Babe, I’m Gonna Leave You

As some of you may have guessed from the subtitle, and the post i general, I have decided to give it a GO in another company. I have come to an agreement with Mærsk Line IT, to start as a Senior Business Intelligence Developer in August. My primary focus areas will be training and guiding new Business Intelligence developers on the SQL Stack, delivering proof of concepts, establishing a baseline of best practices for Business Intelligence development. Besides that, I will engage in the ongoing projects as a developer and try to maintain or up my engagement in the SQL Server Community.
This position also requires some traveling, since some of the projects are done by a Business Partner in India. Traveling, as a part of a job, is new to me, but I’m very excited to go explore the possibilities that lies within.
I am really looking forward starting in the new position. But first – three (3) weeks of vacation 🙂


MsBIP Meeting – Inside xVelocity

On Thursday 26th I will be attending the 5th Microsoft BI Professionals Denmark meeting at Microsoft in Hellerup. The meeting will have prominent guests as Marco Russo (blog|twitter) and Alberto Ferrari (blog|twitter) has offered to stop by and share some knowledge on these two subjects:

Inside xVelocity (VertiPaq)
PowerPivot and BISM Tabular models in Analysis Services share a great columnar-based database engine called xVelocity in-memory analytics engine (VertiPaq). If you want to improve performance and optimize memory used, you have to understand some basic principles about how this engine works, how data is compressed, and how you can design a data model for better optimization. Prepare yourself to change your mind. xVelocity optimization techniques might seem counterintuitive and are absolutely different from OLAP and SQL ones!

Choosing between Tabular and Multidimensional
You have a new project and you have to make an important decision upfront. Should you use Tabular or Multidimensional? It is not easy to answer, because sometime there is a clear choice, but most of the times both decisions might be correct, at least at the beginning. In this session we’ll help you making an informed decision, correctly evaluating pros and cons of each one according to common scenarios, considering both short-term and long-term consequences of your choice.

This session should be loads of fun 🙂


Perculiar Named Calculations

In regards to this article about named calculations in SSAS, I had the perculiar finding today, that NULLs are handled somewhat differently than I would have expected.
In my case I had constructed a named calculation, just as in the article, by concatenating two columns, both of type string and allowing NULLs.
The perculiar part was, that when the second value was NULL, the whole expression would result in a NULL.
So, in order for the concatenation to succeed, the expression should be as follows:

[Marital Status] + CASE WHEN LEN([Has Children]) > 0 THEN  ‘ ‘ + [Has Children] ELSE ” END



Reporting Services og lokal Domain Controller issue


Jeg havde forleden den førnøjelse, at installere en Domain Controller lokalt på det udviklingsmiljø vi arbejder på. Vi har måtte troppe op med egen server, da kunden ikke kunne afse serverkraft til et udviklingsmiljø. Så maskinen står altså i et domæne vi ikke rigtig kan kontrolere.

Reporting Services, og i særdeleshed rsreportserver.config, håndterer det ikke så elegant som man kunne have ønsket sig. Servicen kører, i vores tilfælde, under kontoen Network Services, som erstattes i det øjeblik Domain Controlleren bliver installeret. Kontoen får altså tildelt en ny SID.

Problemet opstår fordi Reporting Services registrerer SID’en på Network Services, som jo erstattes. Hvorfor den nye konto ikke længere har adgang.


Giv kontoen Network Services rediger/modify rettigheder til mapperne: C:Program FilesMicrosoft SQL ServerMSRS10.MSSQLSERVERReporting ServicesLogFiles og C:Program FilesMicrosoft SQL ServerMSRS10.MSSQLSERVERReporting ServicesRSTempFiles


70-452 PRO: Designing a Business Intelligence Infrastructure Using Microsoft SQL Server 2008

Jeg tænkte jeg lige vil opsummere hvad jeg kan give videre til andre, der måtte tage denne certificering.

Inden testen startede blev jeg præsenteret for et indledende spørgeskema hvor man skal svare på hvordan man ser sig selv som professionel i barnchen. Her spørges om erfaring med produktet og i branchen generelt. Det har man ca. 15 minutter til at gennemføre hvilket skulle være rigeligt for de fleste. Hvilke valg man foretager skulle efter sigende ikke have indflydelse på hvilke spørgsmål der kommer i selve testen.

Selve testen er overordnet indelt i flere kategorier end jeg lige havde regnet med. Efter at have taget 70-448 havde jeg forventet samme struktur. PRO testen har, ud over de seks traditionelle kategorier: Administrering/Implementering af hhv. SSIS, SSAS og SSRS, også fundet rum til spørgsmål om generelle datawarehouse udfordringer og brug af TFS.

Den test jeg fik, bestod af 50 spørgsmål fordelt på følgende måde:

Datawarehousing: 2 spørgsmål
Ét om dimensionel modelering og ét om SDC (Slowly Changing Dimension)

SSIS: 11 spørgsmål
Konfiguration, logning, transaktioner, dimensioner før facts, roundtrip minimering, export, backup strategi og SQL Server agent rettigheder.

SSAS: 19 spørgsmål
Herunder sikkerhed, sourcecontrol, hukommelsesforbrug, MOSS, processering, datamining (x5), MDX, attributrelationer, snowflake.

SSRS: 18 spørgsmål
Opgradering til native mode, subscriptions, portering til ny server, skalering, drilldown, drillthrough, custom code, ReportViewer, filtrering, Excel og perspektiver.


Det burde give en kort introduktion til hvad der er i vente hvis man melder sig til denne certificering. Der er dog mange områder der ikke er blevet berørt af testen jeg fik, så man må endelig ikke tage min erfaring som facitliste.