“A little less blogger nation, a little more action please.”
Folks, I’ve been up to my eyeballs in PowerPivot these days. Which ironically has resulted in less writing about PowerPivot.
Quite simply, the release of PowerPivot v1 to MSDN was like a huge starting gun going off. Suddenly, all that theory became “ok, let’s put it into practice… on 10 different fronts.”
I’ve been on so many exciting phone calls and web meetings lately. Lots of very cool customer and partner opportunities.
Oh, and tons of “real” work, too. Which has left less time for blogging.
But I do love the nifty community we’ve got going here and will never neglect it for long stretches. So, while I burn the candle at all three ends, I thought I’d share some quick observations and experiences, rapid-fire style:
- Microblogging has its place – when I don’t have time for a full blog post, I tend to drop something into my Twitter account. If you’re not a Twitter person, you can see what I mean here. And if you are a “tweeter",” well, you can click there, too 🙂
- I fit 200 Million rows, a 63 GB SQL database, into a 1.3 GB PowerPivot Workbook. Yeah, no kidding. 50x compression. And some of the queries against the resulting workbook ran in less than 3 seconds.
- Another db with 120 Million rows exceeded the 2 GB file size limit – different data set. See? There really is no set rule of thumb for how much compression to expect.
- Removing irrelevant columns can dramatically shrink your file – this is well documented elsewhere, but removing columns from your data before importing them can have a HUGE impact on file size. It can be surprisingly non-linear, too – remove one column out of ten and file shrinks by 50%.
- Seems like deleting columns after the fact does NOT shrink the file by as much – I have not verified this scientifically. But I *seem* to get better file size savings by never importing columns in the first place, as opposed to deleting them post-import.
- Early response to PivotStream’s new offering on the PowerPivot platform has been VERY positive – yesterday we demo’d a new set of interactive dashboards, built on PowerPivot, to an existing customer who currently is served by one of our other technologies. It got a big thumbs up, we are moving ahead with a full conversion. And another customer is likely just a few days away from making a switch from their existing on-premise solution, too.
- In general, MS might not need viral adoption – when we were building the product we often talked about how bottom-up adoption by Excel users was crucial, that top-down adoption would be slow. So far, I am seeing zero indication of that. Everywhere I go, everywhere I demo PowerPivot, the organization I’m talking to does a collective “yes, please.” Maybe MS could have charged for the client addin after all. Shhh, don’t tell them.
- I’m seeing more willingness than in the past to embrace the new Office wave – normally, the new wave of Office products takes years to seep into corporations. That will still be true in a lot of places. But about 2/3 of the places I go, I find eager willingness to roll out the new products, even if only on a limited, departmental basis, in order to get PowerPivot.
- Any serious work on PowerPivot measures requires Notepad++ – seriously, I don’t know how I ever got by without it.
- Running a PowerPivot server farm is hard work – I have newfound respect for IT pros everywhere. Don’t go it alone, folks. We’ve brought some heavyweight talents on board to help out with SharePoint, farm architecture, and Integration Services. I can’t imagine where we’d be without them.