The PASS Summit tends to be one of  Microsoft’s favorite venues for unveiling big news in the BI space.  As you may recall, the 2010 Summit revealed some amazing things for the PowerPivot world.

Yes, I know that the 2011 Summit was weeks ago, and I’m overdue on my observations.  And no, I did not attend in person this year.  But the keynote tends to be the vehicle for the big news, and it was available via streaming.  So I watched it later the same day.

It’s a couple hours long, vast stretches of it are dry wooden rhetoric, you can’t really fast forward it, and I don’t recommend watching the whole thing even though the highlights were worth it.  I’ll share those here to the best of my ability.

Point 1:  Denali Release Date “First Half of 2012”

OK, this means we will get the final production version of PowerPivot v2, the new Tabular BISM, and Crescent in first half of 2012.  I was kinda expecting them to say first quarter of 2012, so I was a little surprised.  I guess this means there is still time to get real feedback submitted Smile

Points 2-4:  Cloud and Big Data

A very distinctly “cutting edge” feel to this year’s keynote.  And honestly, there appears to be substance to it, not merely hype.  I would say that the SQL team is one of the most nimble orgs at Microsoft, and one of the most responsive to changing customer needs.

2) “The cloud world is a hybrid of your data center and the cloud”
SQL VP Ted Kummert


This was a very deliberate and prominent statement.  It’s very interesting (and encouraging) that they said this – a sharp contrast to the MS reputation of “our offerings are the only things in the universe and are only designed to work with themselves.”  (BTW – that reputation, while deserved, derives from the academic mindset of MS employees rather than from arrogance, but at times that’s a fine line).

The meaning here is that we will be able to opt in “a la carte” rather than being forced to convert completely to Azure in order to make use of service X.  I like that a lot, because I expect some services to mature faster than others.

OK, maybe that’s not a big deal.  That’s just good business strategy and perhaps obvious.  But there’s a big difference between them stating this as a prominent theme (as they did) versus mentioning it as a detail, or merely bringing it up in Q&A (which is often the case).  They were NOT saying this last year.  So I call this a very positive development.

3) “Reporting Services Will Be Available in Azure Sometime Next Year”
(…and then nothing was said about Analysis Services)

I forget who said this – it was either Ted or Amir, or maybe both.

The real information for me here was what was NOT said.  They said nothing about Analysis Services (SSAS), and the omission simply cannot be an oversight.  It was too obvious, the void in the next sentence was tangible.

That means they either already know that it will be 2013, or they are trying for 2012 but aren’t sure enough yet to promise it.  Either way, we can safely assume we won’t see SSAS Azure until late 2012 at the earliest.

Since PowerPivot is built on Analysis Services, that also means we won’t see any PowerPivot in the cloud until late 2012 at the earliest.  Furthermore, Office 365 won’t support PowerPivot until late 2012, or probably 2013.  That’s not a fact, but it’s a very safe guess.

4) Hadoop Support in PowerPivot!

PowerPivot and Hadoop:  Sounds Like Chocolate and Peanut ButterDo you use Hadoop?  I don’t either, at least not yet, but a number of our clients at Pivotstream do.  So my ears definitely perked up when they said that we will soon have an ODBC driver that connects directly to Hadoop sources.  And as a bonus, our boy Denny Lee got some stage time giving the demo.

Seems like a natural fit – PowerPivot’s ability to crunch large volumes of data coming together with the world’s most popular system for collecting massive amounts of web data.  And again, a departure from the MS norm.  I would typically expect MS to hastily invent a Hadoop competitor and rush it to market, then take five years to make it a credible competitor.  Maybe that’s still a long term goal, but to embrace something with open source and Google roots like this so prominently is again a very novel and mature move that we should salute.

I’m actually getting a more in-depth demo and update today, so I hope to report back with more detail soon.

5) Introducing Data Explorer!


How often do we get something 100% brand new?  Data Explorer allows you to take basically any collection of data sources – like an Excel file on my desktop, a sales data set in SQL Azure, and a demographics data source on DataMarket – and mash them together into a single table.

Even better, it then allows me to publish the resulting data set, in Azure, so that others can consume it.

I have a LOT of questions about this new offering, but very little time to explore it.  I have asked a member of the Data Explorer product team if I can interview them on the blog.  If that doesn’t work out, maybe one of you out there would like to investigate it and submit a review to the blog.

Point 6:  Crescent is now named Power View


Just like PowerPivot was known as Gemini until late 2009, we knew Crescent would eventually get a real name.  And that real name is Power View.  Yes, the space is official.

Point 6a:  Live interactivity in PowerPoint (yes, the slides app) is going to be included in the Denali release after seeming like it was going to get cut.  Pretty cool.

Point 6b:  Purely my opinion, but Power View seems aimed at putting a more glamorous face on traditional BI scenarios – it’s a very “field list oriented” tool which in my experience means that only “data people” will take to it initially.

But I also DO believe that as Excel pros get more and more comfortable with publishing PowerPivot models to SharePoint, they will start opportunistically exploring what Power View can do for them, since Power View can be connected directly to a PowerPivot model and used as an alternative front end (or complement to) Excel Services.