When I’m talking to somebody about a particular topic, I see warning flags whenever I hear things like “win-win” or “no drawbacks.” Most often, that means somebody’s trying to sell me something, or they haven’t thought enough about the challenges at hand. My conversation with Wayne, on the other hand, was a fresh, frank discussion of the pros and cons of several topics.

The Pros and Cons of Democracy

To Americans, the term “President” is synonymous with power because the Commander-in-Chief has the entire Armed Forces at his beck and call. The PASS President does have a volunteer army, but they’re not exactly at his beck and call. They’re volunteers with their own agenda and their own ideas. Wayne can’t simply come up with a vision, pick up the red phone, and force thousands of people to execute tasks.

Wayne observed that database administrators like control – and if we don’t have it, we fight to get it. I loved that observation, and I think it’s right on. That gave me a new appreciation for the difficulty of coordinating an association of volunteer DBAs! After reflecting about it for a couple of days, I think that’s a double-edged sword. While we might be used to fighting for control and winning it, we’re not quite as used to coming up with grand visions and executing on them. That’s more the domain of architects, BI consultants, and entrepreneurs, and that might serve to explain why those people have been more likely to run for Board of Directors positions and win.

In a democracy, anybody can voice their ideas, and PASS hasn’t had a problem coming up with new ideas. Even Twitter played into our democracy discussion. During the Thursday keynote by Dr. David DeWitt, PASS attendees were excitedly bantering on Twitter about the quality of the presentation. Folks just couldn’t get enough of DeWitt’s visual explanations of row storage versus columnar storage. One thing led to another, and people started asking whether the keynote would be included on the PASS DVDs. Rick Heiges, seated next to Wayne, watched the action unfold on Twitter. He turned to Wayne, asked if PASS could get the keynote onto the DVD, and Wayne went to the technical folks to find out. All this happened DURING the keynote! Immediately after the keynote, Wayne took the stage to announce that DeWitt’s presentation wasn’t originally scheduled to be on the DVD, but it would be added due to popular request. Great win for instant interactivity and democracy.

Not all ideas are so easy to act upon, though, because most ideas require money. For example, Wayne would love to do a speaker’s bureau to help DBAs deliver better presentations, but how do we fund it? This problem is compounded by PASS’s recent explosive growth. PASS has grown more in the past 2 years than the previous 8, with many more chapters, yet our funding hasn’t kept pace. Wayne worried that we may have grown ourselves into poverty – grown beyond our ability to effectively fund everything we’re trying to do. Wayne said the single most important thing on his agenda was diversifying PASS’s revenue streams.

PASS Chapters and National Revenue Streams

One way to fix it is to coordinate with vendors to get national sponsorships. Right now, individual chapters approach vendors directly and get small sponsorships, but this isn’t always effective. Small chapters and new chapters don’t have the network of contacts or the time to work with every vendor to get the money they need. I’ve seen this problem firsthand at Quest – it’s painful on the vendor side too, dealing with dozens or hundreds of chapters with constantly rotating leaders. One idea in front of the Board is to get vendors to sign big checks, and distribute that money to the chapters.

Wayne’s first concern was to make sure that existing chapter-vendor relationships aren’t cannibalized. Wayne wants to recognize and continue what’s already working. This isn’t about stealing money from other chapters, but making it easier for chapters that aren’t doing it now. It might be a zero-sum game, and we can’t have the vendors saying, “I’m already spending $X for chapter sponsorships individually – I’ll just move that over to PASS nationally.”

From the vendor’s side, we have to be able to get a consistent message out nationally. If a vendor sponsors PASS nationally, they may want a consistent slide deck with vendor logos on the bottom or on the last slide. In the past, this has been difficult – speakers tend to bring their own slide decks. Chapter leaders have enough on their plates already without trying to strong-arm their speakers.

The Importance of Leaders and Consistency

Over and over during our talk, Wayne kept emphasizing the importance of the chapter leaders. He feels that there’s too much focus on the Board of Directors, and not enough focus on all the chapter leaders who do so much work. The Board, the President, and HQ are all just infrastructure components. He says the single most important thing PASS does as a whole is facilitate conversations between SQL Server professionals. It’s all about the chapter leader who brings good people together to meet once a month and have conversations with their friends. When someone needs a job, they need to be able to pick up the phone and call this network of friends they’ve built.

Unfortunately, though, PASS chapters are often built on the backs of heroic individuals. One person knocks themselves out to do everything – and it’s a lot of work building up a local chapter. When this one person tires out, moves away, or changes careers, the local chapter can be thrown into turmoil if someone doesn’t take up the reins with equal passion.

The PASS Board of Directors has a similar challenge. In the past, as new Board members get elected, there’s been some challenges with handing over duties and procedures from one Board member to the next. Some Board members have arrived only to hear, “Good luck with that!” as the old Board member leaves. In the last couple of years, Wayne and the Board have worked to establish procedures that make it easier to keep things on track as volunteer duties change. (I talked with Executive VP Bill Graziano about this as well, and my interview with him will come next.)

Like any other multi-million-dollar business, PASS requires a lot of day-to-day work to keep the lights on. Accounting, marketing, IT, and even just answering the phones – these tasks are all best done by permanent employees, not volunteers. Wayne puts it best when he says that all PASS members have day jobs that involve SQL Server. To perform these duties, PASS retains a management company with its own set of employees. This group is loosely called PASS HQ. Wayne sees their role as maintaining consistency between boards, and I agree that this group of permanent staff is especially important. When we’ve got a constantly rotating set of people at the top (the Board) and a constantly rotating set of people underneath (the chapter leaders), we need some glue to make sure ideas and plans stick.

Side note – this is where I get a little fuzzy – I’m not sure that HQ as mid-level management is the right vision for PASS. That’s not Wayne’s vision, mind you, I’m just struggling to figure out how I see HQ fitting in. It might help to have something as simple as an org chart for PASS so everyone can understand their role. I think people run for the PASS Board because they’ve done a great job of helping at the local level, and they want to contribute more, but there’s not a way to do that right now. You either lead a local chapter, lead a virtual chapter, or you run for the Board. I think there’s another tier of volunteering that we need to figure out, but I don’t know what it is yet.

Wayne’s Own Leadership

I asked Wayne what he was most proud of from his past year as PASS President. He said Steve Jones asked a similar question during the open Ask the Board session, and Wayne had originally struggled for an answer. Due to PASS’s inherent challenges (a democracy of DBAs), it’s difficult to come up with a vision and push your ideas across to others. He said that in the last year, he’s been the face of the community. People hunt him down to talk to him about their thoughts and ideas for the community at large. All day long at the PASS Summit, he’s been a central point of contact. (I can relate to that – I don’t think I walked twenty feet at the Summit without stopping to talk to somebody. There’s just so many great people with great ideas.)

I would suggest, though, that it’s much easier to get the community excited and aligned if you have a blog.

You can’t get people excited about your ideas if they can’t FIND your ideas.

If you want to be a truly effective leader in a volunteer organization, you need to have a platform where you can communicate your goals and your progress. Someone in your audience will find the idea mesmerizing and say, “I can help make that happen!” I’m even more of a believer in this after the PASS Virtualization Virtual Chapter held our first PASS Summit breakfast event. Dylan Finney, Iain Knight, and Ken Simmons did all the hard work coordinating with our sponsor, VMware, our shirt vendors, and PASS HQ to build an awesome event. We had great turnout, lots of good questions from the audience, and a productive dialog. Personally, I didn’t do jack – I just laid out where I wanted to go, and these three guys carried the entire community there. Kudos and thanks to these guys, big time.

That’s the power of transparent communication, and that’s the one thing I think we’re lacking in the Board right now. From talking to Wayne and other Board members, I don’t have any qualms about what PASS has been up to lately. The more that we open PASS up, though, the more volunteers we’ll get, and the better decisions we’ll make.

In my next PASS Summit post, I’ll recap my interview with Bill Graziano, the current Executive VP of Marketing.

Hamster? I Hardly Know Her!

There’s a few things you never want to hear your doctor or your database administrator say:

• “Wow, I’ve never seen a case that bad.”
• “You mind if I take a few pictures to show at the next convention?”
• “I’ll be right back.  I have to put on gloves and a mask.”
• “Gerbils?  How did those get in there?”

Here’s a few tips to help you break the news easier.

Start with “Yes.” Don’t start fights by disagreeing – instead, agree with them and guide them towards something else.  Here’s a few ways to get started:

• “Yes, you’re not alone there – I’ve seen that conclusion a lot.”
• “Yes, that was a best practice for quite a while.”
• “Yes, that approach works in many scenarios.”
• “Yes, I too believe that I’m all-powerful and bulletproof.”

Identify with the people who made a mistake. I try not to blame someone for a mistake, even if it’s grievous, without including myself in that same group.  Put yourself in the other person’s shoes, and think of a time when you did something similar.  Even if you can’t imagine anybody being dumb enough to, say, drop the production database, odds are you did it once too.  My favorite lines are:

• “I used to do this all the time too when I got started, and I learned the hard way.”
• “You wouldn’t believe how often I’ve seen this same problem at other companies.”
• “It’s not your fault – it’s the tool.  It shouldn’t allow people to accidentally do this, and we all do it sooner or later.”
• “I’ve been through this before – have I ever shown you my scar?”

If you have to point, point at roles, not people. Maybe that developer just started last week, and he inherited a pile of nasty orphaned code from some nincompoop who got fired.  Maybe an outside consultant set up that database.  Maybe a third party vendor wrote code that – no, wait, we never write bad code.  Anyway, don’t point your finger at a person specifically unless you’re holding proof that they did it.  Even then, hold off, because they may have done it at the advice of someone else.  Your goal is to fix the problem, not get somebody fired, so use phrases like:

• “I know this wouldn’t have been your choice if you designed this.”
• “I can totally understand why you’re so frustrated, and this would make any developer frustrated.”
• “Report writers take this approach to queries all the time – this isn’t unusual.”
• “This is a very typical injury for people in your field of work.”

Give everybody the benefit of the doubt. You weren’t born with the incredible experience, skills, and sharp clothes that you’ve got now.  When you popped out of Mama’s babymaker, you started sticking your toes in your mouth.  Go easy on the new kids and break them in gently.

I was not present for the next 30 minutes of Dell keynote, but I did not miss it.

</rimshot>

I’m starting this blog entry now, but don’t refresh for a few minutes.  This is not the keynote you’re looking for.

9:18AM – TGIF – Thank God It’s Finished.  Dr. David DeWitt, Technical Fellow at Microsoft, is taking the stage to talk through a deck called “From 1 to 1000 MIPS.”

9:20 – “Tweet if you think SQL* would be a better name than SQL Server 2008 R2 Parallel Data Warehouse Edition.”

9:22 – He’s diving in extremely deeply based on attendee feedback.  If only Dell would listen that well.

9:23 – DeWitt says database system specialization is inevitable over the next 10 years.  This is evidenced by Azure, the traditional engine, and Project Madison having different paths in different directions.  Our skills can be used across them, but there’s different ways they work.  He’s making a disclaimer, though, that this is an academic talk, not a product chalk talk for Microsoft.  It doesn’t indicate future direction.

Dr David DeWitt

9:25 – Taking us back through time to 1980 when computers had 1 MIPS and 1Kb of memory.  Today, the typical RDBMS has 1,000-2,000x the performance based on the components, plus 10,000x faster storage.  In 1985, the fastest relational databases were running 10-100 transactions per second.  Today, SQL Server on a modest Intel box can do 25,000 TPS (TPC-B), but we need 330 drives in order to keep 1 processor busy.  CPUs and disk are totally out of whack in terms of performance.

9:30 – Disk capacity has gone 10,000 faster over the last 30 years, but if you consider the metric “transfer bandwidth per byte,” things are totally different.  Think in terms of towns with a water tower – they have a ton of capacity, but if they try to get that water through a garden hose, they’re screwed.  That’s where we are with disk.

9:32 – In 1980, sequential transfers through storage were 5 times faster than random transfers.  In 2009, sequential is now 33 times faster than random.  That means you can’t afford to move the heads.  It forces databases to avoid random access and use clustered indexes wherever possible in order to get the best performance.

9:33 – In 1980, it took systems 6-10 processor cycles to access memory.  In 2009, a typical Core 2 Duo takes 200 cycles to access memory.

9:36 – DeWitt is covering the impact on DBMS performance of various bottlenecks like resource stalls, memory stalls, computation, and branch-misprediction stalls on scan-bound queries and join-bound queries.  I want a pre-con from this guy next year.  He’s explaining that databases end up working around L1 and L2 cache stalls.  For non-sysadmins out there, L1/L2 cache is memory located on the CPU itself.  It’s fairly expensive, and there’s not much of it on each CPU.  When you’re buying SQL Server CPUs, this number does make a difference.

Sysadmin Slide Pr0n

9:37 – DeWitt’s slides are just fantastic.  I can’t possibly due them justice with explanations here – it’s a great example of how a picture (or a chart) is worth a thousand words.  He’s explaining how data migrates from disk to memory to L2 cache to L1 cache.  Best animations ever.  He clearly put a LOT of time into these slides, and the value to the attendees is enormous.  I guarantee almost everyone in the audience is learning something.

9:44 – Until now in the presentation, everything has focused on row-based storage databases.  He’s about to transition into columnar databases like what’s going on behind the scenes in PowerPivot in Excel 2010, which explains why it works so well on commodity hardware like laptops and desktops.

Column Store Tables

9:48 – DeWitt’s explaining how columnar databases can pull more data through L2 and L1 cache with each single page read off the disk.  A single disk IO can pull thousands of values from an 8k page for a particular index instead of just a handful.  More fantastic slides.

9:52 – Storing tables in columns can dramatically reduce execution times of some data warehouse queries, but SELECT * is never faster, since you need all of the fields on every row anyway.  The less fields you grab, the more columnar storage pays off.  This design improves CPU performance – that’s a little tricky, though, because it really means the CPU will be busy MORE.  It won’t be idling around waiting for data anymore.

9:55 – “You all are just out there tweeting anyway.”  Talked about his history as a professor – he banned laptops because people were just surfing the web.  These days, however, the students are out there tweeting and conversing about the material live.

9:56 – He’s explaining the different representation models you can use for columnar data storage.  Kind of esoteric for most of us, but it’s really interesting to see how it’s going to work under the hood.

10:01 – One model of columnar storage doesn’t even store row IDs, just column values.  I’m not doing this justice in short sentences, but the database engine itself is responsible for figuring out which values belong to which records.  You can’t update this, because it’s stored in order, but it’s fast as hell.  Perfect for read-only data warehouses.

10:03 – Since CPUs have gotten 1,000x faster and disk is only 65x faster, it makes sense to do a ton of compression on these types of warehouses.  He uses a typical rule of thumb that you’ll get 10x reduction in table size with a column store as opposed to a 3x reduction in a row store.  The engine can use that extra space to store multiple sets of the same data in different sort orders.

10:05 – Explaining run-length expression and how it gives such great compression in columnar databases because so many values are identical.  You can have thousands of customers in the same state, for example, and you can eliminate all of the redundant state records that you would have stored in that column.  Explaining bit-vector encoding, and that one soared over my head.  Explaining dictionary encoding, and that’s near and dear to my heart.  You can take multiple occurrences of long strings (city, name, item, etc) and replace them with a couple of bits.  Commonly used when data isn’t sorted in that order.

10:08 – Column-stored databases aren’t all unicorns and rainbows.  Table scans are particularly nasty, and he’s covering techniques that can be used to mitigate the performance penalties.  I love this kind of academic talk because he’s not saying, “Look, here’s a weakness of Parallel Data Warehouse Edition, and here’s all the ways we thought of working around it,” but you can read between the lines.  It’s jaw-droppingly impressive to see the amount of thought that goes into the storage engine.

10:14 – He’s still explaining ways they can work around execution plan weaknesses to get the best performance.  I’m not covering it here because it’s pretty narrowly focused, and doesn’t have a real-world impact on 99% of us yet.  Grant Fritchey, however, is paying close attention for his next query plan tuning book, no doubt.

10:16 – Watching this keynote, it’s crystal-clear to me why Project Madison (Parallel Data Warehouse Supercalifragalisticexpialidotious Edition) makes sense for data warehousing (big tables, infrequent loads) as opposed to OLTP systems.

Key point takeaway slides:

Key Points 1

Key Points 2

Key Points 3

Standard Edition: Now with Backup Compression

SQL Server 2008 introduced backup compression, but it was only available in Enterprise Edition.  At the time, Enterprise Edition cost around $20,000 more per processor than Standard Edition, so companies couldn’t justify upgrading to Enterprise Edition just to get backup compression.  Companies had to need Enterprise for multiple features in order to stomach the price.  If all a DBA needed was compression, they could buy backup compression software much cheaper than the price of Enterprise Edition.

In SQL 2008 R2, even Standard Edition gets backup compression.  That’s a game-changer, and I’d expect to see smaller companies that do backup compression – and nothing else – to start falling by the wayside.

In addition, Standard can now be a managed instance – it can be managed by some of the slick multi-server-management tools coming down the pike like the Utility Control Point (read my SQL 2008 R2 Utility review).  It can’t be the management server itself – it can’t be a Utility Control Point – but at least we can manage Standard.  It’s good to see that Microsoft recognizes all servers need to be managed, not just the expensive ones.  Big thumbs up there.

Enterprise Edition: CPU Limits

In Enterprise, Microsoft giveth and Microsoft taketh away.  SQL 2008 R2’s BI tools include a new Master Data Services tool.  It’s targeted at enterprises with data warehouses that need to manage incoming data from lots of different sources, and that data isn’t always clean or correct.  MDS helps make sure data follows business rules.  This isn’t a common need for OLTP systems, so it’s only included in Enterprise, not Standard.  Makes sense.

A little less easy to stomach, however, is a new set of caps on Enterprise Edition.  The current SQL 2008 comparison page shows that Enterprise has no licensing limit on memory or the number of CPU sockets.  SQL 2008 R2 Enterprise Edition is capped at 8 CPU sockets, and there’s a memory cap as well, but I haven’t been able to track down a public page showing the cap.  The only hint is the SQL 2008 R2 edition comparison page, which notes that Datacenter Edition (more on that in a second) is licensed for “memory limits up to OS maximum.”  If that wasn’t a unique selling point, it shouldn’t be included in the feature list.

The more expensive Enterprise can act as the management server (Utility Control Point) for up to 25 instances.  However, that doesn’t mean you need to buy one Enterprise per 24-25 Standard servers, and then manage them in pools – there’s an app edition for that.

Datacenter Edition: For, Well, Datacenters

The new Datacenter Edition picks up where Enterprise now runs out of gas.  It supports more than 8 sockets, up to 256 cores, and all the memory you can afford.  Or can’t afford, for that matter.

If you’re going to manage over 25 instances with the Utility Control Point stuff, Datacenter Edition can manage “more than 25 instances” according to Microsoft’s edition comparison page.  I like how they worded that – they didn’t say “unlimited instances,” because there will be performance impacts associated with using Utility Control Points.  The performance data collections gather a lot of data, and storing it for hundreds of instances will take some pretty high performance hardware.

Parallel Data Warehouse Edition: Sold with Hardware Only

The big new fella in town getting all the press is the artist formerly known as Project Madison, formerly known as DATAllegro.  It’s a scale-out data warehouse appliance, but you won’t find this appliance at Home Depot.  This version of SQL Server is sold in reference architecture hardware packages from Bull, Dell, HP, EMC, and IBM.  Write one check, and you get a complete soup-to-nuts data warehouse storage engine that includes everything from the servers, SAN, configuration, and training.

I had the chance to talk with Microsoft’s Val Fontama, and I’ll post more details of that interview next week, but I have to share one quick snippet.  I asked what happens when a Parallel Data Warehouse system starts to have performance issues, and he explained that the DBA will need to call in specialized Parallel engineers.  You won’t be popping open this rack and installing another drawer of hard drives yourself or adding additional commodity hardware boxes to scale out your datacenter.  It’s more of a sealed solution than something you have to build yourself.

I have mixed feelings about this – as a guy who loves hardware, I want to dive under the hood.  However, as a guy who’s managed data warehouses, I know it’s one heck of an ugly skillset to learn on the job, and when data gets into the 5-10 terabyte range, you can’t afford to make configuration mistakes.

How Much Would You Pay For All This?

Ginsu. Accept No Substitutes.

It slices.  It dices.  And if you call now, you can get all this for the low, low sticker price of:

• Standard Edition – $7,499 per processor (socket)
• Enterprise Edition – $28,749 per processor
• Datacenter Edition – $57,498 per processor
• Parallel Data Warehouse Edition – $57,498 per processor (but you’ll be buying this in combination with hardware anyway)

Eagle-eyed readers will note it’s about a 20% price increase from SQL Server 2008.  That’s probably easy to justify on Standard Edition because Microsoft can say they’re throwing in backup compression, a feature that normally would have cost extra from third party vendors.

SQL 2008 R2 Enterprise Edition, however, won’t have quite as easy of a time justifying its price increase given that it now has CPU caps and already had backup compression anyway.

8:42AM – Rushabh Mehta, the next PASS President, is getting things started by talking about the 2009 financials by saying, “Let’s talk about why you didn’t get free drinks.”  You can log into sqlpass.org/governance to review the past revenue & expenses and the future budgets.

8:44 – Projected revenues in FY 2010 are $3.2 million, a 15% drop in revenues, while doing a 40% increase in community spending.  67% reduction in IT expenses.  The European PASSCamp is projected to be profitable on its own.

8:48 – Rushabh is encouraging attendees to speak at local events and volunteer for your local chapter.

8:50 – Wayne Snyder coming onstage to announce the PASSion award winner for this year.  He asked for Tim Ford (Blog – @SQLAgentMan) and he’s not in the room, hahaha.

8:54 – Now recognizing Grant Fritchey (Blog – @GFritchey), who’s wearing a skirt kilt, and Amy Lewis, but I can’t see if she’s wearing a skirt from here.  Recognizing Jacob Sebastian, the PASS Outreach Program leader in India who’s helped start 5-6 chapters.

8:59 – The PASSion award is broken up into 2 categories this year. The International winner is Charley Hanania (@CharleyHanania)!  Congrats, man.  You totally deserve it!

9:01 – The North American PASSion Award winner is Allen Kinsel (@SQLInsaneO)!  He had to review over 150 PowerPoints as part of his work on the Program Committee.  What a great pick.

9:06 – Tom Casey of Microsoft coming onstage.  He’ll be doing a Q&A afterwards through Twitter.  Address your questions to @MS_SQL_Server.

9:11 – Tom’s recapping the MS Information Platform Vision slide that they recapped last year and yesterday to bring folks up to speed.  Initially I cringed, but given yesterday’s stat that 42% of attendees are first-timers, it makes sense.

9:13 – Tom’s selling optimism for DBAs: we need to be optimistic about our future career prospects because BI will be adopted by more users, and we can manage it with the skills we already have.  He’s bullish on the gains caused by commodity hardware, yet saying our skills won’t be commodities.

9:15 – Introducing Ron VanZanten of First Premier to tell his success story.  They consider their 25 terabytes of customer data helps differentiate their business.  They’re having to find new income streams, and their BI stack was the engine where it all comes from.  Every single employee (of their few thousand) gets some kind of information from their data warehouse.  Nice.

9:19 – With Madison, some of First Premier’s hour-long queries run in under a minute.

9:23 – Tom’s discussing a similar theme from last year’s keynote – the skunkworks BI projects.  Superhero power users go grab the data they need without IT’s help.  They grab a connection string and make magic happen with Excel (or Access, but Tom didn’t say that.)  Sorry I don’t have more frequent updates here – it’s just all marketing stuff you don’t really want to read.

9:29 – Amir Netz of Microsoft coming onstage to demo PowerPivot.  Querying millions of rows in Excel and it returns super-fast even on a laptop.  The demo is using a 133 meg Excel file.  It’s stored compressed, and it’s sliced and diced in memory in columnar format.

9:35 – Amir’s adding his own user data via a linked table.  Whenever he updates his Excel data and his source data, they all hold hands and walk down the beach with happy music in the background.  Always live data.

9:38 – Amir’s explaining joins to us.  Joins.  To database administrators.  On the bright side, he’s also showing us a new language we have to learn in Excel.  “All these things that you’re used to doing in MDX, you can now do in Excel!”  Of course, the language is different, but…

9:41 – Demoing Excel 2010’s new slicers feature.  It lets Excel power users create dashboard-style reports that are extremely interactive and easy to use.  Part of me is glad I didn’t dive too deeply into SSRS, because I’d much rather play with this.  Very cool stuff.

9:44 – Demoing SharePoint 2010’s new PowerPivot gallery.  Extremely attractive report gallery.  Tom’s walking over to “his office” – a different desktop, a Win7 touchscreen machine.  Very attractive.

9:45 – The touchscreen Win7 machine doesn’t have Excel, only a browser.  Showing how you get beautiful Silverlight user experience of flipping through a Rolodex of reports like an iPod scrolling through reports.  Looks like it was Bogdan, one of the guys behind the cloud-based data mining add-in for Excel.  I gotta buy that guy a beer.

9:48 – Tom did a great segue into explaining why we need to focus on the experience of interacting with our data.  That demo was extremely captivating – and our reports can be too.  This is the key to making SQL Server sexy to C-level executives.

9:49 – Showing how IT can manage the data refreshes.  When the documents are moved into SharePoint 2010 in the gallery, the IT staff can control when the workbooks refresh their source data.  I’m very interested to see how this scales – IT folks don’t want to click one at a time per PowerPivot document.  We need to schedule documents to refresh sequentially so they don’t all overwhelm the servers at once.

9:53 – Showing motion bubble charts.  I love this stuff.  Extremely attractive.  It works like Google Analytics’ motion charts.

9:55 – The detail of a particular PowerPivot report shows who wrote it, what data sources they used, AND, more interestingly, who’s USING it.  You can see who else finds a particular report interesting, and interact with them.  Making BI more social.

9:56 – Everything shown requires Office 2010, SharePoint 2010, SQL Server 2008 R2.  That’s quite a leap for implementation.  I know a lot of people still struggling on old versions of Office.

9:57 – Demo over.  Today’s keynote flowed much better than yesterday, especially since it focused on telling a couple of stories start to finish.

10:01 – Showing customer-submitted PowerPivot reports. They look pretty impressive.  They look pretty, period.

10:04 – Tom’s announcing a PowerPivot Tweet-to-Win contest for an XBox.  Follow @PowerPivot and tweet this:

RT@powerpivot: Want to learn more, go to www.powerpivot.com and sign up for CTP#powerpivot

10:06 – First time in 10 years they’ve simultaneously shipped a version of Office and SQL Server.

10:07 – Tom wrapped up by saying it’s up to us to build new experiences with our data.  To some extent that’s true, but it’s also up to the C-level executives who will authorize us to implement Office 2010 on the client side.  That’s one heck of a project in big companies.

All done!  Meeting with Andy Warren next….

Maple Bacon Lollipops, aka Man Bait

I ran across these at Whole Foods and immediately bought the entire box.  Erika thought I was crazy.  (Well, I mean, to some extent she’s thought that for years, but this kinda reinforced her case.)  They’re all-natural maple-syrup-flavored lollipops with flecks of bacon inside.

BACON. IN A LOLLIPOP.

When strangers want to lure grown men into vans, this is what they use – as evidenced by the laptop’s name, “Man Bait.”  I’ll give you a moment to wash that thought out of your mind.

I taped my Moo Cards to the back, and I used these as bait to entice SQL Server professionals to race to the SQLServerPedia booth.  My evil knows no bounds.

You can buy your own maple bacon lollipops at Amazon for less than a buck apiece.  This has to be the cheapest bribes you’ll ever use.

You can refresh this page in your browser whenever you’d like to see my latest notes with timestamps.

Opening Remarks by Wayne Snyder

Wayne Snyder is the current President of the PASS Board of Directors. In January, he’ll be replaced by Rushabh Mehta, the current Executive VP of Finance.  Here’s the new PASS Board lineup.

8:00 AM – Wayne’s explaining how things used to be, and how connecting, sharing and learning is different in the age of social networking.

8:01 AM – Summit registrations (not incl. pre/post-conferences) down from 2,445 in 2008 to 2,200 in 2009.  PASS’s goal was to be down just 15% from last year’s registrations, but they’re only down 9%.  Congratulating Bill Graziano and the PASS marketing team for maintaining momentum.

8:02 – 42% of attendees are first-timers.

8:03 – 98 SQL Server MVPs are here.

Wayne Snyder

8:04 – Wayne is thanking Jorge Segarra (Blog – @SQLChicken) for his mapping of the user groups.

8:06 – “Our regional events are always without conflict.”  Hmm – not sure what that’s supposed to mean.  I’m probably reading too much into that.  PASS is going to fund and promote as many local and regional events as possible, even when they’re not official PASS events (like SQLSaturday.)

8:07 – 24 Hours of PASS had 50,123 session registrations from 3,524 people from 70+ countries.

8:08 – Relaunching SQLServer Standard as PASS’s flagship publication as a free area of premium content on SQLPASS.org.  First article of 2009 posted this week, all back issues will be available online.  Encouraging people to submit content and get paid for it.  (For the record, I’m nervous about a user group paying people to write content for a free magazine – that doesn’t scale well.)

8:11 – Come meet the board members for an open Q&A on Wednesday in room 6E from 4:30pm to 6:15pm.  Will be moderated by Joe Webb.

8:14 – You can get your free copy of SQL Server 2008 Standard Edition at the Product Pavilion.  Don’t forget that, really valuable benefit.

Live Twitter Wall with Brightkite

8:18 – Microsoft’s Bob Muglia taking over the stage.  He’s telling the story of Microsoft’s original SQL Server announcement back in 1988.

Bob Muglia

8:21 – Microsoft just handed us a press release.  SQL 2008 R2 will have a November CTP, and there’s two new editions coming – Datacenter Edition and Parallel Data Warehouse Edition (formerly Project Madison.)  PDWE (as I shall name her) will be available on hardware from Bull, Dell, HP, EMC, and IBM.  Not clear who’s providing SAN gear versus hardware – curious to see who’s using which pieces.  Datacenter Edition (above Enterprise Edition) supports up to 256 logical processors and unlimited virtualization.

8:25 – Bob’s showing SQL 2008 R2 on IBM x3950 M2’s with 192 CPUs.  Custom demo screen lets him spin up more or less CPUs – jumping from 64 to 128 cores.  I’ve managed the old x460s that did similar daisy-chaining to get tons of CPU power out of a single box, and while I’m really impressed with the technical achievement, I’m nowhere near as impressed with the cost practicality.  I’ll blog more about that later.

8:29 – Announcing that they set a benchmark world record for TPC-E of 2,012 tspE, plus a record data warehouse on Windows TPC-H 3TB 102,778 QphH.  He’s emphasizing that SQL Server can handle almost anything.

8:31 – Bob’s saying it’s a transformative time for IT because new workers are used to using Twitter, finding things on the web, and using social computing.  We as data professionals need to empower them much more effectively.  I think he’s about to announce Microsoft is acquiring Tweet-SQL.  (I’m kidding.)

8:34 – If you want more info about the IBM servers that daisy-chain together, here’s an IBM PDF whitepaper on the older version of the servers running SQL Server 2005.  It’s got info on the NUMA setup and how to scale SQL Server with it.  Keep in mind that these servers were usually deployed with single-core CPUs, so scaling was a little different back then.

8:38 – Edwin Yuen, Senior Product Manager of the MS Virtualization Team is taking the stage to show Live Migration in Windows 2008 R2 Hyper-V.

Edwin Yuen Demos Live Migration

8:41 – Live Migration is pretty unimpressive if you’ve been using VMware ESX for the last few years, but it starts getting more impressive when they throw in phrases like “move to the cloud.”  Just dropping hints though.

8:43 – Bob retook the stage, and he’s talking about private clouds. I’m psyched for the long-term impact of this if it adds more than virtualization.

8:46 – They want to provide a coherent programming environment across all platforms – your datacenter, a virtualized datacenter, private cloud, and public cloud.  Bob says your SQL skills will continue to pay off.

8:48 – They’re learning things with SQL Azure that will pay off for SQL Server later.  For example, the Data-Tier Application (DAC PACK) is focused on the cloud first, and SQL Server second.

8:50 – You need to transform your role into data professionals – helping the business make decisions with data, not just being custodians of it.  Get jiggy with your data.  (My words, not Bob’s.)

8:51 – Bob’s handing it over to Ted Kummert, the Senior VP for all business platforms (apps plus SQL Server.)

Ted Kummert

8:54 – Ted’s explaining how all the team’s hard work is coming to fruition in the next 6 months with the release of so many things – Azure, the new Parallel Data Warehouse Edition, R2, etc.

8:56 – Ted’s top 5 reasons of why it’s great to be at #SQLPASS 2009:

1. You’re part of the world’s largest gathering of SQL Server professionals.
2. You can take your questions to the source – MS employees, product team, SQLCAT team, etc.
3. Wayne Snyder and Rushabh Mehta.
4. You can work hard and play hard – Microsoft’s GameWorks party should be great.
5. You will build skills & knowledge on the #1 database in the world.  (Unit leader in database market worldwide, fastest growing RDBMS, leaders wave from Forrester, magic quadrant for Gartner for data warehouse & BI, etc.)

9:02 – Cheerleading over, talking about the next year of releases.  Okay, no, wait, more cheerleading – this time about needing less and less fixes in each service pack.  They now have standalone service pack uninstall – says that’s the most wanted feature since the release of v6.5.  I must be out of touch with what features people want.  DBAs, TEST YOUR SERVICE PACKS IN DEV AND QA FIRST.  I’m not playing.

9:07 – Bringing on a customer, First American Title, to talk about SQL Server.  Now’s my chance to run to the bathroom.  Dear God, when will I learn to whiz before the first day’s keynote?

9:12 – I’m back.  The IBM demo machines behind the stage must have been left running the demos, because the fans have become deafening.  That says a lot in a room this big.

9:18 – Dan Jones is onstage to demo multi-server management.

Dan Jones and Ted Kummert

9:22 – They’re demoing the same things I showed in a SQL 2008 R2 video tutorial back in August on SQLServerPedia, except now it’s feature-complete.  Looks great, but they’re now talking about shrinking files as “freeing up space for other applications.” Noooo!

9:28 – Demoing Visual Studio’s ability to define rules about where a DAC Pack gets deployed.  For example, you could require clustering or x64 servers.  The challenge with this for DBAs is that in order to see what’s going on inside the DAC Pack, you’ll need to bust out Visual Studio.

9:31 – Woohoo! The latest build of Visual Studio creates DAC packs that you can apply to existing deployed applications, and upgrade their schema.  For example, in Visual Studio you can add a middle name to an existing schema, and generate a new DAC pack.  Hand that DAC pack file to a client’s DBA, and it will automagically look at their schema and upgrade it to the new one – no schema comparison required, no guessing on what the schema last looked like.  Much easier app upgrades.

9:33 – Bob’s back onstage talking about developer-focused features around the Entity Framework, BI-focused features like consuming data feeds from SharePoint, and platform-focused features like interoperability with other languages.

9:34 – Pablo’s onstage to demo DAC packs and the Entity Data Model in Visual Studio Beta 2 with the .NET Framework 4.  I’m not gonna lie to you – it’s flying way over my head. Or around my interests, something like that.  Couldn’t care less.  I’m not seeing excitement on the Twitter feed either, so I’m guessing I’m not the only one.  When Pablo says things like, “You can either build the app or the data model first,” it makes my ears hurt.

Pablo Demoing Visual Studio

9:42 – I’m still kinda stunned that they’re spending so much time on Visual Studio at the day 1 PASS keynote.  If you had any doubt about Microsoft’s focus on developers, you see it now.

9:45 – StreamInsight will release in SQL 2008 R2, and the programming interface will show up in Visual Studio.  Talking about how McLaren uses StreamInsight to pull performance data off Formula 1 cars.  Now THAT would have made a compelling demo.

9:49 – Ted’s back and talking about pervasive insight – end user access to BI.  Touching on Master Data Services in SQL Server 2008 R2.

9:51 – They’re demoing Master Data Services in a web browser.

9:58 – Demoing scale-out loads of a 10 terabyte data warehouse with 60 billion rows.  They loaded 60 million additional rows in a matter of seconds, did aggregations of several billion in a matter of seconds, etc.  Just looks gorgeous.

Demoing Scale-Out Loads with Parallel Edition

10:00 – Showing visualizations off the massive data warehouse using Report Builder with sparkline graphs and bar graphs.  Reports running very quickly off a 20-node, 336-core system.

10:03 – Ted’s talking about Azure. Says it’s not a new platform – it’s a new method of delivery for an existing platform.  Azure is feature-complete as of today.

10:08 – Azure demo starting onstage, but everybody in the crowd is leaving.  Sessions start at 10:15, and everybody needs seats in the sessions because they’re first-come-first-serve.  Even some of the bloggers left.

10:09 – Showing SQL Azure Data Sync – sorta like 2-way replication between your database and the cloud.  Uses a background agent to manage change.

10:13 – Demoing connectivity between Visual Studio and Azure, so you can deploy your DAC Packs from within Visual Studio.  (Not SQL Server Management Studio – think about that for a second.  Deploy databases as a developer without getting DBAs involved.)

10:14 – Ted hints that Parallel Data Warehouse Edition will be delivered later than SQL 2008 R2.  Makes sense, because there’s hardware vendors involved.  Looks like you have to buy it in a package from the hardware vendor.  I’ll ask more about that.

That’s all folks!  Off to the SQLServerPedia booth.

With no further ado, here’s the best of the blogs this year:

Best Blog Series (Multiple Posts):

• Winner: Kendal Van Dyke for Disk Performance Hands-On Series (Blog – RSS – @SQLDBA)
• Runner-Up: Tie:
  Aaron Alton for SSMS Series (Blog – RSS – @AaronTheHobt)
  Aaron Bertrand for Bad Habits to Kick Series (Blog – RSS – @AaronBertrand)

Best Business Intelligence Blog Post:

• Winner: Chris Webb for Implementing SSRS Drilldowns (Blog – RSS)
• Runner-Up: Michelle Ufford for Using Variables in SSIS (Blog – RSS – @SQLFool)

Best New Blog:

• Winner (Tie):
  Aaron Alton for The Hobt (Blog – RSS – @AaronTheHobt)
  Michelle Ufford for SQLFool (Blog – RSS – @SQLFool)
• Runner-Up: Jorge Segarra for SQLChicken (Blog – RSS – @SQLChicken)

Best Professional Development Blog Post:

• Winner: Kendal Van Dyke for Off-Hours Work Guide for Managers (Blog – RSS – @SQLDBA)
• Runner-Up: Sean McCown for Landing That Job (Blog – RSS – @MidnightDBA)

Best T-SQL Blog Post:

• Winner: Aaron Bertrand for Stored Procedures Best Practices Checklist (Blog – RSS – @AaronBertrand)
• Runner-Up: Grant Fritchey for Unpacking the View (Blog – RSS – @GFritchey)

Best Unusual Blog Post:

• Winner: Alex Kuznetsov for Reproducing Deadlocks with One Table (Blog – RSS)
• Runner-Up: John Magnabosco for Encrypting Large Values (Blog – RSS)

We’re going to put together a badge for the winners to display on their blogs, and we’ll be doing some email interviews with the winners. If you’re at #SQLPASS this week, congratulate these bloggers for the killer work they’ve put out, and if you’re not, take a moment to subscribe to their blogs and send them a congratulatory tweet. They’re doing this just for the love of the community, helping out folks like you and me, and we need to recognize them for working so hard.

I’m going to be in the SQLServerPedia booth until 5pm, but then I’ll be joining the Q&A session after 5.

If you can’t make it to the live Q&A session, you can write down questions and put them in a suggestion box in the PASS booth.

The PASS Board is made up of volunteers who want to serve the community, and they were elected by the community.  Please keep that in mind when you ask questions.  Don’t use accusing tones, don’t refer to them as “The Man,” don’t spout off about how they’re trying to keep you down, and please, for the love of all that’s holy, don’t ask why so many of the BoD members are bald.  It’s just not polite.  Besides, they look great bald, especially when they tattoo their heads.

• Getting the Last Clean DBCC CHECKDB Date
• Tracking Password Changes
• Server Audit Examples

A group of SQL Server DBAs and bloggers voted the DBCC CHECKDB date article as the Wiki Article of the Month!

The winner of the Article of the Month gets perks big time. They get a Box O’ Swag, and they’re invited to join the Quest Experts Community. The Experts get licensing & support for the Quest SQL Server portfolio (LiteSpeed, Spotlight, Performance Analysis, Capacity Manager, Change Director, Toad, etc.), early downloads of betas, inside access to Product Managers, and more. (Coincidentally, Microsoft MVPs get this same access – if you’re an MVP and you didn’t know about this, go register for the community now.)

If you’d like to see your name in lights, check out the How to Contribute page at SQLServerPedia. If you’re not sure what to write about, check out the article requests by section. These are topics our readers have found interesting or would like to learn about. When you’ve picked a topic, shoot me an email and I’d be glad to help you get started. You can use material you’ve written for other sites, too, like your own blog.