Tag: Tabular

Categories
Business Intelligence

Day 3 and Overall Impressions #sqlpass #summit12

To conclude my posts on the PASS Summit this year (see day 1 review and day 2 review), I want to go over the last day and then talk a little about the entire conference and my takeaways.

On Friday I attended three sessions. The last two of the day, the second to last one I was on the phone and missed the beginning and then decided to check out all the other things before they were done, and the last session slot, there wasn’t much appealing, and most everyone already left, so I skipped it.

1. CLD-303-A SQLCAT: What are The Largest Azure Projects in the World?

This was given by Kevin Cox and a another SQLCAT member. The SQLCAT team is crazy smart. If you can talk to any of them, you need to. Any chance you get. This was a good view of customers they have dealt with that are pushing SQL Azure to the limits. Since we are running a project now that we are going to be pushing SQL Azure (and Azure) hard, I thought this was good.

So customers have 20TB dbs, 10k databases, so it for sure can scale. Also some good tips/tricks on what you can do to use SQL Azure to the max like the other customers

2. BIA-203: Real-Time Data Warehouse and Reporting Solutions

I wasn’t sure on this one. Carlos Bossy gave a couple of presentations, and he seemed to know what he was presenting, but a topic like this is so situational it is tough to make it generic. Also, there isn’t a “huge” need for real time and I think I wouldn’t implement it the way he was saying anyways. Run SSIS 24/7 with a for loop that never ends? That is crazy. I’d rather pump data through something like StreamInsight with some code than that SSIS solution. Or run things every couple of minutes or something. “near real time”. Also his solutions was using replication which is fragile.

3. BIA-402-M: Optimizing Your BI Semantic Model for Performance and Scale.

Probably one of the better sessions. Again, Microsoft guys letting it all out here. Akshai Mirchandani and Allan Folting from Microsoft. Basically going *in depth* on how PowerPivot and Tabular does what it does with columnar compression, etc. Where you can look and dig under the hood to find ways to make small changes and optimize processing or querying depending on your need. This is a session I want my entire BI team to watch together.

Overall Takeaways Technically:

Azure, Hadoop, Tabular, Power View, BI, DAX, Excel. You can see a pattern here. I am sure there were good “DBA” and DB Dev sessions but I didn’t go to any. BI is taking shape with Microsoft’s strategy and it is all tabular/excel azure/hadoop stuff. Exciting times.

Overall Thoughts of This Year’s Summit and SQL PASS

Guidebook – was ok. I thought it could have been better. I have used before at conferences. Why no native Windows Phone app?

New Layout – the last two years I was at the summit, things were laid out (as far as where things were) pretty much the same. This year it was changed up. Took a day to get the “lay of the land”

Keynotes – kind of the same as usual. I mentioned in by part 2 blog about how the blogger/twitter table needs to grow up, just want to say that again here. First day there was some drama, second day more drama and badmouthing/infighting. Just needs to stop. Leave the drama at home.

Seattle – Seattle is great. Not going to Seattle next year since the summit is in Charlotte, is going to be tough. I know where to go in Seattle and I like the area. I am worried out Charlotte.

Reg Dates – as I mentioned in my day 1 review, many people came out a day early since the dates said 6th-9th. Same thing next year. Should really say 7th-9th.

Hash Tags – on twitter, usually the hash tag is #sqlpass .. this year they said use #summit12 , some people were using #summit2012 and confused. Also, using #summit12 wasn’t looked at by as many people which stinks as I used that on all my tweets. Next year they need to just keep it as one hash tag.

Karaoke – I have been to the unsanctioned one. It was great. Not sure sanctioning karaoke like this year makes sense. It loses some of what made it cool to begin with. I could get into a lot of detail here but I hope people understand what I mean… taking something “underground” and trying to make it mainstream, usually doesn’t work as well.

#sqlfamily – this is something that I have many thoughts on. I will say things but I don’t think many want to hear it. “sqlfamily” isn’t as big of a family as those in the echo chamber think it is. I would say 99% come to the summit and have no real idea of what it even means. 1% that tweet, present, schmooze think everyone else feels and interacts the same way they do, and it just isn’t true. I met many people at breakfast/lunch and after hours that in fact have no real want/need to be totally ingrained with the clique. Many don’t even use twitter, etc. They are just going to work, doing their job, trying to learn. etc. I think it would make sense fo the sql/sql pass community to step back and think about that for a while.

This year I wrote a blog post for the SQL Server Blog before the summit to drive excitement, which was cool. The first day at the keynote a guy sat next to me and we were talking before it started. He was like “dude, I read your blog on the sql server blog!” – To me that was so cool. He said I was a “rockstar”. No, I am not a rockstar (or an MVP – but the blog says I am, maybe the emails have been going to my spam folder all these years) – I am just a regular tech guy that is passionate about technology, SQL, BI (and a ton more). I was really happy though to see that people are reading that content and it is firing them up, it is what my intention was. And if you read that post, I took back a ton of good stuff from the summit. I am already starting to formalize and get strategy/implementation plans going for things I directly learned.

So to close, my third summit was great. Great content, meeting new people and seeing old faces and having lively discussions and knowledge sharing during the day and over a beer. I am going to miss Seattle next year but I can’t wait for the next summit, and possibly even the new SQL BA (Business Analytics) conference in April 2013. I hope everyone who went to the Summit this year enjoyed it and learned as much as I did!

Categories
Business Intelligence

Day 2 Review #sqlpass #summit12

To follow up on my first post about day one of this years PASS Summit, here is how day two played out

The “keynote” here was some PASS discussions, then Quentin Clark (MSFT exec) and Julie Strauss (wicked smart) doing an end to end demo on many things.. Hadoop, Azure, Data Explorer, Power View, Excel, etc. The blogger table was pretty annoying with their tweets during the demo calling it out as boring and not what DBA’s want, failing to remember that half the conference is BI people. I think the demo was “dry” but they showed many things and tied it together. I saw Julie at TechEd and she knows what she is doing. Of course every year the blogger table is going to say “zoom” on the presentations, which yes, they should be doing, or changing resolution, but to see the bantering back and forth on twitter is just bad overall for the people attending and watching and looking for info. The blogger/twitter table should be relaying information that people at home are clamoring for, not bad mouthing the presentation/presenters.

I hit up 4 sessions in all on Thursday Nov 8th..

1. BID-307-M: Using Power View with Multidimensional Models

As with day one, I mentioned I try to get to presentations by Microsoft employees, today was no different. The first one being with Bob Meyers and Sivakumar Harinath. This was a deep dive into the newly announced functionality yet to be released or given a date that will let us hit OLAP cubes with Power View. Honestly I wish Microsoft would have released this from the get go. One thing I don’t understand though is why Power View uses DAX to hit OLAP and TABULAR, while Excel uses MDX to hit OLAP and TABULAR. Seems split brained to me. Choose one and go. Many audience questions in this one, and one downfall of Microsoft Employee presentations is that they have a hard time saying “no” and get into discussions with audience members, many times taking too much time on some specific question.

Presentation was good, and we learned some things. New dimension properties for ImageUrl, Geography (for mapping), etc. And what will and won’t work with Power View and OLAP. Good stuff.

2. BIA-400-HD: Enterprise Data Mining with SQL Server

This was a double session, and I just stayed for the first half. Mark Tabladillo (marktab) is a PhD so that tells you something. Data Mining in SSAS/SQL Server has always been an enigma since day one. I don’t know of many using it in real life (besides the AdventureWorks Demo?) – it is kind of SSAS Cube Writeback, awesome, but not widely used. He showed how you can use the SSAS Data Mining cubes and Excel Add in to do forecasting, basket analysis and how to get into some of the options and get data out yourself to make your own visualizations, pretty cool stuff, but like I said, I left half way through…

3. BIA-309-M: Enriching Your BI Semantic Tabular Models with DAX

I left the Data Mining session early to get a good seat for this presentation. Kasper de Jonge from Microsoft is one I always try to get to as he is wicked smart as well, and usually the presentations are awesome, this one was no different. Getting into the details with DAX and just seeing someone like Kasper use PowerPivot, Excel .. it shows how “he” would use it, being a program manager, which is different than most. Great to pick up tips/tricks and just see how he goes about doing even the basics. He even showed off the trick on changing the DAX on an imported table to a DAX query to get whatever you want back from your tabular cube, he has a blog post that I went through a while ago to the same effect, which was cool.

4. BIA-206-M: BI Power Hour

Finally to end the day..Matt Masson and Matthew Roche again, with Patrick LeBlanc, Peter Myers, Sean Boon and Chuck Heinzelman.

This presentation reminded me of a Brian Knight spectacular.. throwing trinkets, books, etc to audience, goofy stuff. Pretty funny, and they go through SharePoint, SSIS, PowerView etc. Very lighthearted and a good way to end a 2nd day on non-stop technical things. Matt Masson is probably a stand up comedian at night, just funny stuff. I have seen Chuck present before and he is good, Sean showed us some PowerPivot with Olympic data and Shark bite data, Patrick with a Windows Phone app and Azure and SQL Data Sync, Matt with SSIS data app, and Peter Myers filled in at the end by capturing data from the audience over mobile and slicing/dicing it. I have seen Peter before as well and he is very methodical, it was his first “power hour” and it showed, but hopefully he does it again and is a bit more prepared.

Thursday night was the appreciation night, and gather at the EMP (music museum) in Seattle. They shuttle you over and back. Two free drinks, food (I think I had mac and cheese 3 nights in a row for some reason last week), and you can tour around the museum. There was #SQLKaraoke, but the sanctioned one, not the one at Busch Gardens. Live band and you get to sing, pretty cool stage and everything. Again, bummer, my voice was out or I would have sang a tune.

So to wrap up my 2nd full day, BI, BI, BI all day. More to come with the last day and overall thoughts for this year.

Categories
Business Intelligence SQLServerPedia Syndication

Selling Management on SQL 2012

2012 is going to be a big year in the SQL world. No, the world isn’t going to end. SQL 2012 should get released by Microsoft, hopefully in the first half (cross your fingers for the first quarter!) of the year. Great! But many out there are now on SQL 2005, or 2008, or 2008 R2, some even on SQL 2000 (SP4 – still get support?) but you want to get to SQL 2012. What can you do to make that transition easier? You need to sell the features and benefits, just like anything else.

Clustering

If you have any kind of clustering environment, or mirroring, or are even thinking about doing clustering, then SQL 2012 is going to be what you want to do. With AlwaysOn, it makes it dead simple to create and manage clusters. If you look back over the versions of SQL, and think clustering, you might shutter. With 2012, things become much easier and management has to see this benefit, as with anything, to make your systems more available with the new AlwaysOn

Master Data Services and Data Quality Services

Microsoft came out with their first round of Master Data Services (MDS) in SQL 2008 R2, but it was lackluster. The interface is clunky, weird, and hard to use. Most “end users” of MDS aren’t going to be that technical. You need something simple, like SharePoint, or Excel. MDS is neither (even though its a weird version of SharePoint). With 2012, MDS is vastly improved and actually something viable where an Enterprise could use it for a Master Data Management (MDM) solution. Couple that with Data Quality Services (DQS) and you get tons of bang for your buck. with MDS and the excel add on, this will be just what the doctor ordered for MDM groups in businesses.

Business Intelligence

Near and dear to my heart of course, is Business Intelligence. What a huge release for BI folks in 2012. First off, a whole new analysis services type, Tabular. Columnar Vertipaq type cubes. Reverse engineer PowerPivots right into SSAS Tabular and then tweak to release out to the Enterprise.

Then the enhanced SSIS stuff, better IDE, better management of packages, and more. Of course the integration with the Visual Studio 2010 IDE is a welcome feature, especially for those of us that also need to work on C# and .NET 4.0 stuff!

But don’t forget the potential biggest thing yet out of the BI tools for 2012 – Power View (yes the space is intentional, not sure why.. but now we have PowerPoint, PowerPivot and Power View). Naming aside, Power View could be a HUGE analytics tool to get more BI out to the people in an Enterprise. First off, they plan on making it work on iOS! Power View works on tabular cubes, so you see the tie in there. The one big thing with Power View, is it just works inside of SharePoint. No stand alone editor. You better have SharePoint 2010 and a pretty good SharePoint admin along side your BI team to get all this stuff working. Some of the enhanced end user alerting in SSRS integrated mode looks nice as well. But once again, you need SharePoint! DON’T for get the SharePoint!

There is much more in SQL 2012 that will make DBA’s lives easier, and BI pros development streamlined. Too much to outline in just one post. But if you are trying to sell SQL 2012 upgrade to management, the “big three” things I outlined above are a good starting point. One thing to be aware of though is that the licensing model has changed in SQL 2012 to core based, so you would want to read up on that.

I’m excited for SQL 2012 bits to hit and I hope you are too!

Categories
Business Intelligence

#sqlpass 2011 Summit Thoughts

Want to get a quick post out before I forget everything cool I have been thinking about the 2011 SQL Pass Summit in Seattle WA.

The conference started out great with a first timer reception (this was my second year), and I knew a few first timers so that would great.

The first day keynote was good, and then the sessions I hit were on a full range of things, new semantic search stuff with file table, and more.

I missed a couple of sessions on day 1 because I was in a Microsoft led focus group on “BI in the cloud” some very cool ideas thrown out and excited to see what is coming up in the future.

More good keynote on day two, a deeper dive into the stuff in SQL Server 2012 (formerly Denali). Went to some good sessions on Data Quality Services (DQS) and Vertipaq vs OLAP, and Power View (remember the space!) SharePoint and Power View integration, etc. Some very cool stuff coming soon with the 2012 release.

Tabular cubes vs Multidimensional cubes. OLAP vs Vertipaq, etc. Big debates and questions here, when to put things in Tabular vs MDX, etc. DAX solves some hard problems we run into with MDX.

Another thing I went to a couple sessions on was StreamInsight. This is some very cool technology for complex event processing, using .NET and LINQ. The StreamInsight guys are crazy smart. Also integrating StreamInsight with Azure, and doing some crazy processing and analysis of tons of data in near real time. I see this technology making a difference somewhere down the road.

Power View was another big winner from the summit. This integrates with SharePoint 2010 as a Silverlight (SL still isn’t dead?) report builder and viewer. Not even sure they are called “reports” anymore, but “views”. You create views and you can interact/edit them in real time. Export to PowerPoint too. These views work off “tabular” vertipaq BISM (BI semantic model) cubes. This technology is going to change how we deliver information to the business.

Being able to take a PowerPivot (no space!) and reverse engineer it into SQL Server Data Tools (formerly BIDS – in VS2010 shell) to create a BISM tabular cube and publish to SSAS Tabular instance looks great. Solving problems like crazy snapshotting and many to many and time analysis using DAX looks very useful.

Of course, GameWorks appreciation night was fun, and I found a Mcmenamins (Six Arms) up the road from the convention center, so I got get a taste of Ruby again. Good conversation and times at the Taphouse and around the Sheraton and Convention center was good too. Seattle is a nice city and the weather was pretty good I thought. Another great summit.