On my way home from the PASS Summit in Seattle, I had a layover in Amsterdam before continuing onto Copenhagen. For various reasons, we were about one and a half hours delayed, and I arrived in AMS at 9:30, my CPH flight departing at 9:35. As you’d probably guessed, I missed my flight.

To fix it, I was told to do a check-in at one of the self-service counters at the transfer desk. Apparently these both do normal check-ins, as well as (supposedly), get you a replacement flight in case you missed the normal one. I stuck my passport into the scanner and it gladly popped up with a new flight to Copenhagen about 4 hours later. Great! Well, except my name wasn’t on the check-in list, two other (unknown to me) Danish names were. However, I did share a surname with one of them.

After asking one of the attendants I was told, “Oh, that’s because your flight has already departed, it’ll try to find the nearest match”. Áha, so if there’s no 1:1 match, it’ll just try to find the best match of an existing booking to a non-departed flight – and apparently the best it could find was a two-person ticket for a flight 4 hours later. Now, it was for the same destination, and I did share my surname with one of the passengers, but it surely wasn’t my ticket. Still, I was able to check-in as them and confirm “my” ticket.

So this leaves me the question, what in the world do they (Delta/KLM/AF) use as the primary key for their ticketing system? Once my passport is scanned, and they’ve got my passport number, how can there be any doubt as to who I am? How in the world can I get to check-in two completely, to me, unrelated passengers on a flight I’m not going on myself?

I ended up getting booked for another flight to another (closer to my home) destination in Denmark, but it did require manual intervention from the service desk.

It’s 8:15 AM and I’m back at the bloggers table, ready for the day 2 keynote. The format will be the same as yesterday.

On a side note – I’m surprised at how comfortable wearing a kilt is.

[8:18] Bill Graziano is on stage, cheering at all those of us wearing #sqlkilts. Next up applauding all the volunteers, speakers, chapter leaders, etc. Thank you!

[8:23] Lori Edwards is awarded the 2011 PASSion award, well deserved!

[8:27] Video on stage praising new SQL Server 2012 through personal testimonials. Looking forward to hear about the new features, hopefully non-BI today.

[8:30] Quentin Clark (Corporate VP, SQL Server, Microsoft) on stage. Using the same lovely Metro based slidedeck that we saw yesterday.

[8:31] Urges us to stop using the word Denali and instead use SQL Server 2012. Lots of hardware & appliances on stage with Quentin – appliances will probably be a big topic, just as it was at SQLBits.

[8:32] SQL Azure powered by SQL Server 2012 at this point. 180+ new features, can’t showcase all, we’ll see top 12.

[8:35] Required 9s & Protection: SSIS as a server, HA for StreamInsight, SQL Server AlwaysON – HA/DR being the main keyword.

[8:36] Bob Erickson (Executive VP of Interlink Transport Technologies) on stage telling about their experiences with SQL Server 2012 Mission Critical deployment. Lots of fancy buzz words, no technical details or practical details.

[8:41] AlwaysON being demoed live on stage. Unfortunately fonts and UI is so small, even with scarse zoomit usage, that noone can see what actually goes on.

[8:44] Zoomiit used on stage, bloggers table goes wild in applause.

[8:45] AlwaysON demo is cool and all, but this is the same stuff we’ve seen at multiple presentations already.

[8:47] Quentin has moved onto the next topic – Blazing Fast Performance.

[8:49] Performance enhancements in RDBMs engine, SSAS, SSIS, ColumnStore indexes (previously known as Apollo). ColumnStore indexes are seriously interesting. Any kind of demo would’ve been cool.

[8:50] Quentin moving onto next two topics – Rapid Data Exploration & Managed Self-Service BI. In other words, a rehash of what we saw yesterday. It looks cool, most DBA’s I’ve talked with are somewhat reluctant to letting their users get access to self-service BI.

[8:51] Credible Consistent Data. Lots of buzz words, hard to extract any practical details.

[8:53] Quentin welcoming Lara Rubbelke on stage – taking a jab at the Excel demos done yesterday. Demo starts out with Sharepoint interface with miniscule fonts – as usual, noone can see what’s happening. Sounds like this is going into ColumnStore demo.

[8:56] Lara creating a ColumnStore index live on stage, using Zoomit too. Stil needs to run at way lower resolution for everybody to be able to see what’s happening. Lara ditching the GUI and writing the index in T-SQL instead, lots of cheers from bloggers table.

[9:00] Lara demoing cloud based DQS, Azure marketplace, ColumnStore indexes.

[9:02] Quentin onto next topic – Organization Compliance. Added audit options as well as user - defined server roles.

[9:04] Next topic – Peace of Mind. Production simulated application testing – distributed replay hint. SCOM advisor & management packs. Expanded support – Premier mission critical, no direct details yet.

[9:05] Next topic – Scalable Data Warehousing + Fast Time to Solution. Hardware vendors delivering hardware applications. HW+SW+Support – turn on the faucet, buy appliances, preoptimized. This is the SQLBits keynote. Choice of hardware – semi-appliance option as well?

[9:10] Going through hardware on stage, introducing the Dell & HP appliances.for PDW, DW in general. Plug’n’play appliances. Appliances sound great, but aren’t they just overpriced versions of what you can do yourself? Add virtualization and you’ve got a complete appliance in just a VMDK. From boot to data loading / production usage – 20 minutes. If you have the money, this is a great & quick way to get started.

[9:14] HP appliance can be ordered a month from now. Can start small and scale up – “Don’t know how big you can built it, haven’t reached limit yet”.

[9:19] Next topic – Extent Any Data, Anywhere – ODBC drivers for Linux – unreadable white on light orange background slide text.

[9:21] Beyond relational – adding new support for FileTable – FileStream evolved. 2D spatial, semantic search.

[9:25] Semantic search being demoed on stage by Michael Rys. Semantic search is basically an addition on top of FTS adding more language specific semantic improvements.

[9:28] Next topic – Optimized Productivity – Juneau / SQL Server Data Tools reference. Shipped with SQL Server 2012. Unified across DB & BI, meaning no more SSMS/BI, just one “studio”. Deployment & targeting freedom – just like .NET dev, we can target a certain platform and be limited to just hte options available. Talking contained databases.

[9:29] Last topic – Scale on Demand. AlwaysON, deployment across public & private – hybrid cloud/private reference? Elastic scale = cloud reference, spin up many instances. It’s not vertical scaling, this is sharding through SQL Azure.

[9:31] Nicholas Dritsas on stage demoing deployment through SSMS 2012 / SQL Server Data Tools. “And I know how to get a big applause – Zoomit” – keynote speakers seem to be catching up on the zoomit hype. Thank god!

[9:33] Speakers referencing zoomit means they’re checking the tweets realtime – excellent.

[9:34] This is cool – SSMS being used to manage both local and Azure databases. You can now manage Azure without defaulting to the online web based management portal. This will make it much easier to achieve the “hybrid” approach of having both on-premise and cloud based databases.

[9:35] Now demoing how to do SQL Azure backups to Windows Azure. Lovely, and about time. Why wasn’t this here from the start?

[9:39] Cihan Biyikoglu entering the stage to talk about Elastic Scale – also known as SQL Azure Federations.

[9:43] New Azure management portal uses Metro UI, Looks slick, but does it scale?

[9:45] SQL Azure will now scale up to 150GB. But if only way to do “elastic scaling” is federation, whether it’s 50GB or 150GB doesn’t really change much other than your shard size. Since we’re still limited by unknown hardware, we may not be able to even utilize a 150GB database before having to federate.

[9:47] Execution plans in the Azure web interface based on Metro, looks slick.

I’ve found my seat at the Bloggers Table, ready for the beginning of the keynote in just 12 minutes. I’ll be doing my best keeping up with all of the exciting news about to be spilled out.

Todays keynote will be delivered by Ted Kummert (Senior VP, Business Platform Division, Microsoft) and Rushabh Mehta (PASS president).

If you follow this post, I’ll add new comments at the end of the post, ordered by time. Note that my blog aggresively caches content so you may have to clear your session to see the newest version of the post.

Live content

[8:15] Room’s starting to fill up, there’s a LOT of people – looks like way more than last year. Hoping Rushabh will reveal some number porn for us during the keynote!

[8:20] Light dimming down, PASS love testimonials blasting off on the 4 giant screens next to the stage.

[8:23] If you’re not here, remember you can tune in to the live streaming feed here: http://www.sqlpass.org/summit/2011/Live/LiveStreaming.aspx

[8:25] Rushabh Mehta on stage. Welcome all – THIS is community!

[8:29] Rushabh talking about goals – 430k technical hours out of 1000k goal. 80k members out of 250k goal. 1 global region out of goal of 5. 20k new members alone this year, impressive!

[8:31] The community has grown – 24HOP, SQLSaturday, SQLRally, PASS Chapters, Virtual Chapters & of course the summit. Also actively collaborating with SQLBits & SQLServerFAQ.

[8:32] PASS – largest single event in the world dedicated to SQL Server & technical BI content.

[8:34] 189 sesions, 5 tracks, 204 speakers (93 MVPs, 11 MCMs). Biggest SQL Server event.

[8:39] SQL Server MVP Deep Dives vol II available. Profits going to charity. Get it! Book signing will be happening at the summit.

[8:41] Rushabh welcoming Ted Kummert on stage.

[8:44] “Some database vendors just decided to get into the cloud last week. You know who I’m talking about right?”

[8:46] Over 4,000 registered attendees at the PASS Summit. Largest Summit conference ever!

[8:48] Cloud is not either or, it’s a hybrid world. Some will be in cloud, some will be in your own datacenter. We want to provide choices.

[8:49] Want to offer all the cloud capabilities, SSAS cloud version hinted at.

[8:50] Denali – one of the most significant releases in our history.

[8:51] Crescent will end up as Power View on release.

[8:53] Denali is just around the corner. Official name – SQL Server 2012.

[8:54] Denali / SQL Server 2012 being released in first half of 2012.

[8:55] “Believe in the platform as more than just relational data.”

[8:57] “No one size fits all.” Ted Kummert talking on Big Data.

[8:59] Microsoft supporting the Apache Hadoop project. Make sure it’ll run great on Windows. Is this the belief in “more than just relational data”?

[9:00] Connect Hadoop and Microsoft and offer them together. Microsoft submitting to the Hadoop project to ensure it runs great. “We’re going to connect the dots in our environment.”

[9:01] Ted Kummert welcoming Eric Baldeschwieler (Hortonworks CEO) on stage.

[9:02] Eric looks like he’s about to fall asleep. Should’ve had a Red Bull before going on stage before 4,000 people.

[9:04] “Hadoop could be storing half of the worlds data in five years.”

[9:06] Ted Kummert welcoming Denny Lee on stage (Principal Program Manager, Microsoft). Demo time!

[9:07] “Hadoop for Windows Cluster”.

[9:09] Denny Lee demoing HiveQL querying against Hadoop. Hadoop console 100% compatible with classic Hadoop.

[9:10] Denny Lee showing Excel. Why oh why, I’m here for SQL Server, not Excel. Accessing Hadoop cluster through “Hive/Hadoop ODBC driver” from Excel.

[9:13] CTP of Hadoop on Azure by the end of the year.

[9:14] Always nice with a demo, but really, noone could read console (please use Zoomit next time, don’t use white on black). Rest was just pie charts and Excel with a story of where the data came from.

[9:16] Demo of Data Explorer coming up in just a sec.

[9:17] “Who’s excited to be at PASS today?”. Talking about self service BI – one (yes, ONE) person in the room clapped. That’s a 1:4000 ratio.

[9:19] Ignoring content, it’s obvious that the Metro UI team had a say in the slide & UI design of slide decks and products being demoed. Beautiful.

[9:21] More pie charts. I made my first pie chart back in the 90’s, and I was way late to the game.

[9:22] Oh, Excel just reappeared. What is this, PAE – Professional Association of Excel?

[9:23] Talking about Excel & pie charts. Noone can see the screen, even in front row. Resolution way too high, no usage of Zoomit. You’d think they’d get presenting 101 right for a keynote demo.

[9:29] Not much to update. Demoing at unwatchable resoluition, talking Excel, BI, PowerPivot. It looks cool, but I just don’t believe in production drag’n’drop development/BI.

[9:32] If you ever wondered what a golf clap sounded like, make sure to watch the PASS Summit live stream recordings later today.

[9:36] Amir Netz welcomed on stage. This should be better, hoping for some tech demos. Amir is an excellent demoer, based on previous keynotes.

[9:40] More sales, BI, graphs. At least Amir Netz utilizes Zoomit.

[9:50] Amir showing lots of graphs, doing data mining on movies, actors, etc. Most of todays demos seem to be in building on top of what we saw last year. Hadoop announcement was awesome, rest just seems like slow evolution.

[9:56] Keynote about to end. Lots and lots of Excel, BI and graphs. I saw two interesting announcements:

• Denali will be released in first half of 2012 as SQL Server 2012.
• Microsoft actively supporting the Apache Hadoop project and bringing it to the Windows platform.

[10:00] Now on to the coffee to wake up from the keynote. Looking forward to live blogging the day 2 keynote tomorrow – it can only be better than what we just saw today.

I’m glad to announce that I’m one of the lucky few who’ve been appointed a seat at the coveted bloggers table during the PASS Summit keynotes. This means I’ll have an excellent view of the leynotes, and I’ll be live blogging and tweeting from both.

Wednesday 8:15-10:00 AM, PST:
Ted Kummert, Senior VP, Business Platform Division, Microsoft
Rushabh Mehta, President, PASS

Thursday 8:15-10:00 AM, PST:
Quentin Clark, Corporate VP, SQL Server Database Systems Group, Microsoft
Bill Graziano, Executive VP of Finance, PASS

For both days, 8:15 AM PST will be 4:15 PM in Denmark. The live blogging will be in English, as usual. Once the PASS Summit is over, I’ll be following up with a full summary in Danish.

One of the most important aspects of presenting is to stay within the allotted time slot. You should aim at finishing exactly on time, neither exceeding nor finishing too early.

Some prefer to rely on the room monitor giving you the 15, 10 and 5 minute signs, while others prefer to just watch the clock on the laptop. Personally I prefer neither. Depending on the venue, you may not have a room monitor available. Also, the room monitor might not be as precise as you’d like – giving you 30 min marks, etc. As for using the laptop, that may work well during demos, but while Powerpoint’s running, you’ll have a tough time checking the time.

What I do is to open up my iPad, turn off the auto-locking and maximize the screen brightness. Depending on the style, I’ll then place the iPad in an easily viewable location with a big watch on it. This allows me to always have a clock readily available in an easily readable format. I can also put it on the floor enabling me to walk around, while still being able to get the time by just glancing quickly towards the floor.

Examples

Here I am presenting an all day session (as evidenced by the Red Bulls). From my view I have complete control of the time and audience. The timer on the laptop is the Zoomit break timer.

Also works good lying down to avoid blocking anyone’s view of the screen behind me.

Fits snuggly under the mouse wire.

If you look closely, you’ll see the iPad right in front of me, on the floor, allowing me a perfect time view even when walking around on the stage.

After getting back to Denmark from an excellent SQLBits 9, I’ve had a chance to reflect a bit on my experiences. Not just on SQLBits itself, but the very concept of attending conferences and how to get the most of it.

SQLBits

What can I say, it’s an excellent conference. The price is very low compared to comparable offerings, and there’s even free content if you’re just going on the Saturday. I loved the informality, placing it somewhere between the more serious PASS Summit and the king of informality, Miracle Open World.

One of the really great things about SQLBits was the layout. There was no separate exhibition hall, instead, it was smack in the middle of the lunch and walkthrough area. This meant people constantly went by the exhibitor booths and generally had a ball in the common area.

I presented a precon and a general session and I’m generally satisfied with the outcome. I look forward to receiving the feedback from SQLBits so I know how to create even better sessions next time! A special thanks goes out to Simon Sabin who not only cheered me on to continue OrcaMDF back in April, but also pushed me to turn my internals session into a fully fledged precon.

Now for a couple of observations and tips to get the most from the conferences you attend.

Sit up front

Some sessions were packed, others were scarcely populated. In almost all the sessions, you had a pretty free choice of where to sit. If the screen isn’t too close, and there’s space, make sure to get a front row seat! You’ll get an unobstructed view of the screen, better connection with the speaker and optimal conditions for saying hi to the speaker afterwards.

I used to always place myself near the back, just as I’d preferred it in school. Wouldn’t want that nasty teacher to ask you a question, would we? Well, this isn’t school! You’re most likely here voluntarily, so why not get the most from it? The speaker won’t ask you questions unless you participate yourself, so there’s nothing to fear. However, you should participate!

Participate

If there’s anything you’re in doubt of, disagree with or would like to have explained in further detail – raise your hand! The speaker will politely make it clear if there’s no time for questions, but most likely, he/she’ll enjoy the interaction. And don’t worry, there are no dumb questions – you’ll most likely find others pondering the same question, but just not wanting to raise their hand.

Not only will participation increase the chances of you going home satisfied, it’ll also improve the dynamics of the session. As a speaker, there’s nothing better than the feeling of an engaged audience.

Don’t be a wallflower

Wes Brown wrote an excellent piece on plucking wallflowers. I’ve attended quite a number of conferences so far, but never have I gotten as much out of them as I have this time at SQLBits. Why? I met people!

I went to the PASS Summit last year, not knowing anybody. I had a great time, but I didn’t meet very many people – all to the credit of myself. As most of us are, I’m an introvert – talking to strangers doesn’t come naturally to me. However, I know I should, so I try to do my best. At PASS I signed up for the excellent Inappropriate Sessions Meetup, a small community event where everyone seems to know each other. Not knowing anyone, it was a fun way for me to push some boundaries.

This year at SQLBits, I had the advantage of knowing a couple of people in real life. Even better, I’d connected with 20-30 people over Twitter during the last half year. I hadn’t met any of these outside of Twitter, but this really laid the foundation for a lot of great connections.

When you walk around and spot someone whose name you know, introduce yourself! Doesn’t matter if you don’t have anything clever to say, just shake their hand, introduce yourself and walk on if you feel like it. This makes it much much easier to start a conversation later on in the bar, during a session or lunch, etc.

If you liked a presentation, walk up to the presenter afterwards and introduce yourself. If you don’t have any comments, just say how you liked the session – presenters love that. And once again, you’ve set yourself up for an easier conversation later on.

Speakers aren’t special

On the topic of presenting yourself to people – don’t be afraid of speakers. The amount of speaker bite related injuries have been on a steady decline for the past several years, making it safer than ever. Joking aside, you’ll find that most speakers are just normal persons who love to have a conversation. Though speakers may seem very extrovert, many are actually introverts that have just managed to overcome that social boundary in the very moment. Talk to them! Introduce yourself! Set yourself up for conversation later by shaking their hand after a session.

I had a number of people come up to me during the days and just comment on my session, and all of a sudden, we had a conversation. It’s a great way of starting a conversation, and I love meeting new people, getting feedback on my sessions and hearing about their conference experience.

Stay at the conference hotel

The conference hotel is usually the most expensive option. At the same time, it’s one of the cheapest boosts you can give your conference experience. Staying there makes it much more convenient for you to stay at the bar until 4 AM while making friends and awesome conversation.

It’s also much more convenient when you realize the coffee you’ve spilt on your white shirt during the first session. And even more convenient once you get back with your new shirt and realize you also spilt coffee all over your badge and business cards, and thus have to make yet another trip to the room. I totally did this, and it didn’t take me more than 10 minutes to be back in the game.

Staying at the conference hotel will also make it much easier to hook up for dinner events, going out after the sessions, etc. Basically, it’ll make the tough social game easier – it’ll improve your odds of not becoming a wallflower.

The PASS Summit

I’m leaving for Seattle this Saturday. This’ll be my second PASS Summit, and I absolutely can’t wait. I’m fully determined to get the most from the Summit this year, taking advantage of all the after hours events. There are basically public social events Monday – Friday, many days are even double booked.

Make sure to get the most out of the Summit not just by attending all of the great sessions, but by connecting with people – that’s the one thing you can’t do at home sitting in your chair while watching the DVDs!

I’ve got a blog post coming up tomorrow on general SQLBits reflections (‘twas awesome), for now I’ll just post my session materials as promised :)

Precon attendees

To those of you who attended my precon – thank you! I’m trying to see if I can get the materials sent out to you directly. Until I get that figured out, you can email me at mark@improve.dk and I’ll send the materials directly to you. I’m not keen on publishing the whole slide deck and demo archive so I hope you understand.

Knowing the Internals – Who Needs SQL Server Anyway?

Once again – thanks for attending! It was an extremely fast paced session but I hope you got an idea of what’s going on behind the scenes of SQL Server.

You can download a rar file containing the slide deck, the SNSC tool and the demos here:

sqlbits_2011_09_30_knowing_the_internals.rar

If you saw my session, you might be interested in the following links (besides the internals related posts on this blog):

OrcaMDF on Github
SNSC on Github

Time flies – it’s been about four months since I originally introduced my pet project, OrcaMDF. Since then, quite a lot has happened and OrcaMDF is somewhat more capable than when it started out. As a result I thought I’d provide a recap of what OrcaMDF is currently capable of, as well as what my plans are for the future.

Page types

OrcaMDF currently supports full parsing of Data, Index, TextMix, TextTree, GAM, SGAM, IAM, and PFS pages. It also supports minimal parsing of the Boot page, used for finding the starting point of the base table metadata.

Remaining are Sort, FileHeader, DiffMap and MLMap pages. As MLMap and DiffMap use the same bitmap format as IAM, GAM and SGAM pages, parsing those should be straightforward. FileHeader is a bit more tricky and will require some DBCC PAGE love. Sort pages are less relevant as those are only used temporarily while SQL Server is running and should thus not be stored in your MDF file.

Data types

I’ve been adding more and more data types to OrcaMDF, lately I’ve added support for parsing all of the LOB types except for XML. Currently supported data are:

• bigint
• binary
• bit
• char
• datetime
• decimal
• image
• int
• nchar
• ntext
• nvarchar(x)
• nvarchar(MAX)
• smallint
• sysname
• text
• tinyint
• varbinary(x)
• varbinary(MAX)
• varchar(x)
• varchar(MAX)

Adding support for further data types is relatively easy, it’s just a matter of analyzing the storage format and implementing the ISqlType interface.

Table & index structures

Using the DataScanner class, OrcaMDF is able to scan both clustered tables as well as heaps. Using the IndexScanner class enables you to scan nonclustered indexes, whether they’re applied to a clustered table or a heap.

Metadata

The only publicly exposed metadata that OrcaMDF currently exposes is a list of table names. Internally, OrcaMDF is able to parse both indexes, tables, partitions, allocation units and columns. This enables you to scan a clustered table/heap/index, providing just its name. OrcaMDF will automatically parse the schema of the object and discover where the IAM chain starts (for heaps) or find the root page for indexes.

Notable leftovers

OrcaMDF currently only supports single-file databases, that is, no secondary data files. Adding support would be trivial, but my efforts are concentrated on supporting the core data structures in data files, so secondary files wouldn’t change much. As for corrupted files & corruption detection, OrcaMDF assumes the MDF files to be in perfect condition. There are many places where corruption could be detected, but I’ll postpone that until the day I feel OrcaMDF is able to parse most functioning databases correctly. The projected start out (and still is) as a way of getting a deeper understanding of SQL Server internals, and as such, corruption detection is less important at this stage – though it’s definitely something I’d like to add eventually.

What’s coming next

I want to extend the current metadata parsing capabilities, especially with focus on exposing the metadata publicly. Through OrcaMDF, it should be possible to reproduce the database/object/column tree as seen in SQL Server Management Studio. I want to provide table names, indexes, schemas, keys, etc. I also want to take a look at compression, starting out with row compression. As the format is rather well documented (compared to LOB structures, for example), it shouldn’t pose too many problems.

If you have any suggestions or features you’d love to see supported, do let me know!

I’ve also had a number of requests for examples of how to use OrcaMDF. I’ll probably end up creating a series of blog posts showcasing examples of how to use various features of OrcaMDF through code. I’m also planning to create a post on how to fetch the source code and compile it and finally how to run it from there – just to make it a tad easier to take it for a spin :)

My fall schedule is starting to fall into place, and boy, is it a busy one. I am absolutely thrilled that I’ll be going to the three major SQL Server conferences this fall: SQLBits, PASS Summit and PASS SQLRally Nordic. While I’m thrilled that I’ll be attending, I’m even more thrilled, humbled and excited over the fact that I’ll be presenting at all three.

There’s a lot of people I’m looking forward to meet at this years events, presentations I can’t wait to see, and beer I can’t wait to taste :)

September 6th, Aarhus, Denmark

As preparation for my training day at SQLBits, I’ll be giving a free preview of my presentation. Trifork has kindly offered to provide facilities and refreshments. We’ll start up at 1PM and continue until 7PM. You can register (once again, totally free) here: http://gotocon.com/aarhus-2011/freeevent/index.jsp?eventOID=3556

September 15th, Malmö, Sweden

I will be giving a 60-minute presentation on the undocumented storage internals of SQL Server at the Swedish SQL Server User Group. Registration is free, though limited. Make sure to register before space runs out! - http://sqlug20110915.eventbrite.com/

September 29th, Liverpool, UK

I’ll start out SQLBits by giving a full training day on the SQL Server Storage Engine and MDF File Internals. During this full day session, I’ll give a thorough walkthrough of the MDF file internals, how data is stored, parsed and used optimally. I will be providing a plethora of demos and slides on never-seen-before undocumented internals. I couldn’t ask for a bigger recommendation than what Simon Sabin has written on his blog: Must attend training day for anyone serious about SQL. You can still make the £375.00 SQLBits early bird discount (expires midnight, August 26th) on the complete conference, including the training day.

September 30th, Liverpool, UK

For those not attending my training day, I’ll be giving a 60-minute version where I’ll condense my training day down to the most useful & interesting topics. Hopefully you’ll end up with some valuable knowledge and an unquenchable thirst for more internals knowledge. As mentioned above, you can still make the early bird discount before August 26th – so what are you waiting for?

October 11-14, Seattle, USA

While the final scheduling hasn’t been announced yet, I’ll be giving a lightning talk at this years PASS Summit. In my Revealing the magic lightning talk I will, with the speed of a lightning, attempt to walk you through how to parse the data of a given table given nothing but a table name and an MDF file. We’ll start by parsing the boot page and end up parsing the data records. My goal is not for you to be able to do this afterwards (it’s only 5 minutes, after all), but hopefully I’ll demonstrate that it’s not all magic. Only a little bit.

October 22nd, Aarhus, Denmark

For those of you who prefer to get hands on, I’m hosting a 6 hour code camp on SQL Server internals & performance tuning. I’m aiming for a developer oriented audience who have limited to some experience with SQL Server. A laptop will be required. During the day I’ll be switching between presentations and workshop mode. I’ll introduce concepts after which you’ll be able to try it out yourself, with me helping out as necessary. While it’s aimed at developers, I’ll aim high, don’t expect me to go too easy on you ;) I’ll post once the signup page has been launched.

November 8-9, Aronsborg, Sweden

At the last event of the year, PASS SQLRally Nordic, I’ll be giving my Knowing the Internals, Who Needs SQL Server Anyway? session. This is an improved version of the session I held earlier this year at Miracle Open World. Since then, I’ve spent almost four months developing OrcaMDF, not to mention turning it into an open source project. Suffice to say, I’ve got a lot more to show this time around.

In collaboration with Trifork, I’ll be giving a free preview of my SQL Server Storage Engine and MDF File Internals training day at SQLBits.

The event will take place at Trifork A/S, Margrethepladsen 4, 8000 Aarhus C on the 6th of September from 1PM – 7PM. Note that there’s a limit of just 30 attendees, so make sure to register soon if you want to join.

Click here for more details and free registration

Trifork has kindly chosen to sponsor some sandwiches and drinks for the night, that way you won’t be running completely dry by listening to me for six hours ;)

Whether you’re a SQL Server DBA/Dev that wants to take it to the next step, by gaining an extreme insight into the internals, or just a developer with a passion for bytes & hacking, this should be an interesting session for you :)