Chris Webb's BI Blog

Analysis Services, MDX, PowerPivot, DAX and anything BI-related

PASS Summit 2010 Day 1

with 8 comments

So day one of the PASS Summit is drawing to a close, and what a day it’s been. I did a session myself today, which went well (I think), but the real news is all the cool new stuff that was announced at the keynote and at sessions throughout the day and that’s what I’d like to reflect on here. A lot of questions about what we’ve seen today remain unanswered (more will be revealed tomorrow, I’m told) but I thought I’d blog about what interested me and was relevant to BI.

All in all, I’m very excited. Let’s face it – over the last five years, in SSAS and in the wider BI stack, there’s been a distinct lack of anything really radical and new. But today saw several announcements that will completely change the Microsoft BI stack:

  • There will be column-store indexes, provided by the Vertipaq engine, inside the SQL Server relational database. Simon Sabin has a link to more details here.
  • There’s a new, corporate BI version of PowerPivot, BISM
  • Project Crescent is a new ad hoc query tool coming from the Reporting Services team

Let’s consider what each of these means in turn. Based on the limited information we’ve got so far, column-store indexes in SQL Server massively increase the performance of reporting/OLAP style queries inside the relational engine; if we get column-store indexes in Parallel Data Warehouse, well, I can imagine we’ll get astounding performance over huge data volumes. The pdf linked to in Simon Sabin’s blog says:

Users who were using OLAP systems only to get fast query performance, but who prefer to use the T-
SQL language to write queries, may find they can have one less moving part in their environment,
reducing cost and complexity. Users who like the sophisticated reporting tools, dimensional modeling
capability, forecasting facilities, and decision-support specific query languages that OLAP tools offer can
continue to benefit from them. Moreover, they may now be able to use ROLAP against a columnstore-
indexed SQL Server data warehouse, and meet or exceed the performance they were used to in the past
with OLAP, but save time by eliminating the cube building process.

To paraphrase, if your data’s in SQL Server that’s where it should stay for querying and reporting – MOLAP no longer provides any performance benefit, so the concept of ‘processing’  a cube to get data into a different, OLAP database is gone. If you still want to do OLAP on SQL Server it will be a form of ROLAP, and to me this makes a lot of sense.

Very few details about BISM, the ‘corporate’ version of Powerpivot/Vertipaq, were released but we saw that we could develop BISM models in Visual Studio and the end result was exactly what PowerPivot creates when you deploy a PowerPivot mode to Sharepoint – I guess it’s basically a different type of Analysis Services database. BISM can either hold the data itself in its own Vertipaq store (useful when the data comes from anywhere other than SQL Server, eg Excel, Oracle, text files and so on) or it can act purely as a semantic layer and allow querying data in SQL Server in a ROLAP, multidimensional way. So we’ll be able to create an Excel pivot table, connect to BISM and use that as a thin layer to query data that’s directly in SQL Server. This is clearly what the future of Analysis Services is going to be – while the Analysis Services we know and love today might hang around for a few versions, it’s not got a long-term future in my opinion.

There was a full session on Project Crescent later on today, which I went to, so I have more details on this. It’s essentially a new ad hoc query tool being built by the SSRS team – interestingly it seems to have nothing to do with the rest of Reporting Services, and it doesn’t create rdl files that can be edited in Report Builder or Report Designer. It’s all about flashy visualisation and interactivity and reminds me a lot of Tableau and other fashionable BI tools; I can also see how it can be a replacement for PerformancePoint for creating dashboards. It is only surfaced in Sharepoint (boo! yet another Sharepoint dependency!) and is built in Silverlight; also, it can only work with data sourced from BISM/PowerPivot models. Once you get past the flashy stuff it does much the same that every other ad hoc query tool has been doing since the year dot (Marco, sitting next to me, commented that it was doing much the same thing that Data Analyzer was doing 10 years ago) but the flashy stuff is very flashy indeed, and very impressive – for instance the ability to export views out to PowerPoint slides looks cool; but I do wonder whether it will be as practically useful as something like Tableau so we’ll have to wait and see. I’m pleased to see that someone at MS has finally woken up to the fact that end users might want to use something other than Excel for ad hoc querying.

OK, time for a few drinks and to have some fun! There’ll be another post with more news tomorrow…

Written by Chris Webb

November 10, 2010 at 1:40 am

8 Responses

Subscribe to comments with RSS.

  1. Chris,
    thanks for sharing the news – sounds very interresting to me! And i hope you had some good drinks and fun. I’m curious to read the next news today.
    Dietmar

    Dietmar

    November 10, 2010 at 7:00 am

  2. [...] j’ai apprécié le résumé de la première journée faite par Chris Webb, qui comme François Jehl se concentre sur les nouveautés BI qui [...]

  3. Just wanted to say your session was really great, and I thought you made some really important points concerning the hidden costs of using PowerPivot vs SSAS. You should definitely make your little quiz available online!

    Erika

    November 10, 2010 at 8:01 pm

  4. Thanks for sharing your views on Project Crescent …

    Amit

    November 11, 2010 at 7:44 am

  5. [...] Chris Webb’s PASS Summit 2010 Day One [...]

  6. [...] SQL blogosphere has been lit up with PASS Summit [...]

  7. Project Crescent is very interesting really

    diyer

    December 3, 2010 at 8:53 am

  8. [...] Webb. PASS Summit 2010 Day 1, PASS Summit Day 2, PASS Summit Day 2: The Aftermath, PASS Summit Day [...]


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

Follow

Get every new post delivered to your Inbox.

Join 2,868 other followers

%d bloggers like this: