After the Proclarity acquisition earlier this year, my thinking was that while it was a good thing overall for the Microsoft BI community it would have the negative effect of reducing choice and innovation. But in the months since I’ve been surprised to see a steady stream of new third party applications enter the market. Some, like RSInteract, I’ve mentioned before, but since there are quite a few now that are worthy of mentioning I thought I’d round up a few in a single posting.

Radius, from 90 Degree Software, has got a bit of exposure over the last few days thanks to Russell Christopher’s blog entry. Russell sums up its features pretty well and I have to say I was quite impressed by the various webcasts I’ve seen – I like the idea of being able to reuse pieces of other people’s reports while building your own. I’m not sure it supports AS as a data source at the moment, alas, but I was told that they are planning on doing so soon. While the general rubbishness of Report Builder can only help the prospects of Radius it also highlights the point I want to make here quite well: although Microsoft have now killed the market for traditional, vanilla BI apps, if you are innovative and execute your ideas efficiently there are still going to be openings for you.

Of course visualisation has been another area where client tools have sought to set themselves apart from the pack over the last year. I blogged about Tableau a while ago and Fractal:Edge the other week, and more and more other tools are improving their visualisation capabilities. Take a look at the pie-chart tree report or the data mining report demos at http://www.reportportal.com/ for instance. It’s not surprising that the likes of Dundas (disclosure: I recently signed up to their partner program and got a freebie chess set in the post from them this morning) are also entering this space – Dundas OLAP 5.5 just got released, and has some cool features like the ability to draw freehand over or add comments and arrows to a report; I see Chart FX have something similar, Chart FX OLAP, but I’ve not checked it out yet.

Another way to distinguish your product is to focus on a particular niche. I saw a demo of Intelligencia the other week and initially couldn’t see what it offered beyond any other more general-purpose tool. However it’s the details that are important, and this tool is targetted specifically at financial users who want to create printable reports for regulatory (Sarbanes-Oxley etc) reasons. This is something that you’d struggle to do with, say, Reporting Services, and having the tool as a Word 2007 addin makes a lot of sense; the OLAP querying functionality is also aimed at financial-style reports, making it easy to build up asymmetric sets of tuples for example. Also in the financial apps space is the more general CPM tool Calumo; looks impressive and is probably more sophisticated than Biz#/PerformancePoint will be, and I suppose the advantage of being a small software company in this area is that you can respond quickly to your customers’ requests and try to stay one step ahead of Microsoft.

So, then, a thousand flowers are blooming and all that. Having cool features does not necessarily entail commercial success but I wish all these tools well and I’m pleased to see that there’s still a lot of diversity out there – it can only benefit us all. And remember, if you’ve got a product that uses Analysis Services and you want a bit of free publicity on this blog then please send me details…

It was recently pointed out to me (by Hugo Lopes and Paulo Faria of Enabler in Portugal, who I had the pleasure of working with recently) that setting the ProcessingGroup property of a dimension can have quite a big impact on the amount of time it takes to process a dimension. By default this property is set to By Attribute, which means that each attribute in the dimension is processed separately using its own SQL Select statement which returns the distinct values for each member on the attribute. Here’s an example query used to process an attribute based on the Adventure Works Product dimension:

SELECT
DISTINCT
[dbo_DimProduct].[Color] AS [dbo_DimProductColor0_0]
FROM [dbo].[DimProduct] AS [dbo_DimProduct]

If you set it to By Table, however, in some circumstances (when your dimension is built from a single table?) only one SQL Select statement per table in the dimension structure is issued, reading all the data in, with AS presumably finding the distinct values itself.

When do you use which setting? I suppose it depends on a lot of things (performance of your data source, size and structure of your dimension, whether you’re processing other objects at the same time) so it’s going to be one of those things you’ll have to experiment with. Obviously most dimensions process extremely quickly anyway, but if you’ve got a large dimension which you need to process as quickly as possible this might come in handy.