Category: PASS Community Summit

PASS Programs Team Leverages Data Science for Attendance Predictions and Feedback Rating

December 17, 2014 – This year, Data Scientist and PASS volunteer Dev Nambi joined the Programs team to help with PASS Summit 2014 sessions scheduling and predictive analytics. Over the past month, Dev and the team have been digging into the data from both session attendance and session feedback, and we wanted to share the findings with the community.

In this first year of doing prediction for attendance, we saw a variety of results, including the reduction of overcrowded sessions. Dev details the results and provides a link to the raw dataset in his blog post PASS Summit Attendance and Predictions. With the lessons learned from our debut effort, we will continue to improve on and incorporate prediction models into our Summit scheduling process.

Dev and the team also examined Summit 2014 session evaluations. This year, we saw a 14.8 % return rate of evaluations from attendees, and we’re looking at ways to increase that number. Despite an increased possible margin of error based on the decrease in responses, we appreciate all the session evaluation feedback we received from attendees and are taking key findings into account as we start planning next year’s Summit. To view the correlations and rankings, please see Dev’s analysis in Say Anything! PASS Summit Feedback and Ratings.

A big thank you to Dev for the time and effort he spent analyzing and compiling the attendance prediction and session evaluations data. Thank you also to Summit attendees for using the Schedule Builder to help with attendance predictions and to the entire PASS community for your feedback. Moving forward, PASS Programs will continue assembling, sharing, and using this information and data analysis to help make the Summit experience the best it can be. If you have any comments or input, please don’t hesitate to email program@sqlpass.org.
– Amy Lewis
Director, PASS Programs

PASS Summit 2014 Pre-Con Preview: Brian & Devin Knight

Oct. 16, 2014 – Go inside PASS Summit 2014's full-day pre-conference sessions in this Q&A series with our presenters. We caught up with Brian and Devin Knight for a quick chat about their pre-con, SSIS: Problem, Design, Solution.

Q: If your pre-con had a theme song, what would it be and why?
Devin:
I’d go with Kenny Loggins – “Danger Zone” because it’s an awesome song and so is our pre-con!

Q: What excites you most about working with SQL Server Integration Services?
Brian:
I like the logic problems. Each package you build is like solving a puzzle that has a dozen ways of solving it. 

Q: What's the most surprising statement attendees might hear you say during the pre-con?
Devin: “Okay, now everyone line up for the ceremonial Knight brothers pre-con hug.“

Q: What’s your favorite SSIS transform and why?
Brian:
The Swiss Army knife of any SSIS developer is the Script Transform. When you can’t do something in a native or third-party component, it’s time to whip out a script. We should cover about a dozen different patterns of how to use these in our pre-conference session. 

Q: What still trips you up in the real world when scrubbing duplicate data?
Devin:
Using Regular Expressions to scrub data has always been something that has thrown me, because it feels very unnatural to write coming from a SQL Server background. Luckily, there are plenty of resources such as http://regexlib.com/ to help!

Q: If attendees could start putting into practice just one thing after this pre-con, what would you want that to be?
Devin:
I find that the package restartability section of the pre-con has a lot of content that’s often overlooked but that can be extremely useful for recovering from failed package executions. I think people skip these steps because most people don’t plan for their packages to fail.

Check out our other Q&As with PASS Summit 2014 pre-con speakers.


SQLSaturday at Summit: Q&A Panels, Website Sneak Peek, and More

Oct. 14, 2014  Hello, PASS Community! It seems like it was just a couple months ago that we were meeting up for PASS Summit 2013 – I can’t believe it was actually a year ago. I’m looking forward to seeing you all again in person in Seattle next month.

As part of our Summit agenda, we always dedicate time to meet with our community organizers, including Chapter and Virtual Chapter Leaders, Regional Mentors, and SQLSaturday organizers. As Director of the SQLSaturday portfolio, I wanted to share some of what we’ll be discussing around SQLSaturday on Community Day at Summit, which is Tuesday, Nov. 4, as well as in the Community Zone throughout the week.

Kicking off the annual SQLSaturday Round Table will be a Q&A with the SQLSaturday leadership team, consisting of myself and our passionate Community Evangelists, Karla Landrum and Carmen Buchman. You can review the agenda here, and please bring all your questions and suggestions. We’re following up this year with some activities that go even further in supporting PASS’s mission statement:

Empower data professionals who leverage Microsoft technologies to connect, share, and learn
through networking, knowledge sharing, and peer-based learning.

We'll be hosting a new panel featuring some of our most experienced SQLSaturday organizers from around the world sharing their knowledge and tips and tricks for putting on successful events.

We’ll also be reviewing exciting updates to the SQLSaturday website, coming your way by the end of 2014. Working from your SQLSaturday website wish list – which we’ve cultivated over the past few years – as well as with a Community Focus Group, we’ve taken your feedback and ideas and put a great amount of time and consideration into building a more functional and modern web experience for SQLSaturday organizers and attendees alike. We can’t wait to show off the current version at Summit. You can find a list of all the wish list items that we’re integrating into the next SQLSaturday website here. We’ll also be emailing it along with a full Round Table meeting agenda to all SQLSaturday organizers for review in advance.

In addition, PASS Community Evangelists and SQLSaturday organizers from around the world will be available in the Community Zone Wednesday-Friday to answer all your questions – whether you’re looking for a SQLSaturday to attend or want to learn how to host one yourself. Check out the Community Zone Spotlight schedule to see when to stop by and meet with organizers from your area.

Our team is working hard to make these meetings as informational, transparent, and successful as possible, so we can quickly answer your questions, respond to any issues or concerns, and spend more time discussing how we can improve our future events. As owner of the SQLSaturday portfolio as well as a PASS Director-at-Large, I can’t wait to connect with all our passionate SQLSaturday organizers, volunteers, speakers, and attendees, share what we’re doing today to enhance everyone’s experience, and learn how we can better support SQLSaturday efforts around the world.   

Don’t forget to wear your favorite SQLSaturday shirt Wednesday at Summit, and I’ll see you in Seattle!
– Tim Ford
PASS Director of SQLSaturday


 

PASS Summit 2014 Pre-Con Preview: Allan Hirt

Oct. 7, 2014 – Go inside PASS Summit 2014's full-day pre-conference sessions in this Q&A series with our presenters. We spent a few minutes with Allan Hirt to learn more about what to expect in his pre-con, The A to Z of Availability Groups.

Q: If your pre-con had a theme song, what would it be and why?
Tough choice, but thinking about it, maybe “La Villa Strangiato” by Rush. Besides the obvious if you know me, it’s an instrumental that starts off deceptively simple, but gets really complex and furious at times. I think it’s appropriate. AGs are deceptively simple, but in reality, not at all once you really start looking under the hood.

Q: What excites you most about Availability Groups?
It takes the best bits of a few different things and mashes them into one. That creates certain challenges, but that’s part of the fun.

Q: What's the most surprising statement attendees might hear you say during the pre-con?
If you don’t understand Windows Server failover clusters, you will fail at deploying AGs. 

Q: What’s the biggest myth around AGs that you’d like to debunk?
A secondary replica can be under-powered. That is completely false. 

Q: In the spirit of A-to-Z: What are some first words you’d like to say about AGs? How about a statement to leave attendees with?
First words: The Availability Groups feature is NOT AlwaysOn. AlwaysOn is not a feature.
Last words: Plan your deployments well, and you will be very successful with AGs.

Q: If SQL Server pros could start putting into practice just one thing after this pre-con, what would you want that to be?
I'd like them to understand things like networking, storage, and Windows Server failover clusters so they will have better deployments of AGs.

Check out our other Q&As with PASS Summit 2014 pre-con speakers.


PASS Summit 2014 Pre-Con Preview: Davide Mauri

Oct. 7, 2014 – Go inside PASS Summit 2014's full-day pre-conference sessions in this Q&A series with our presenters. In this Q&A, Davide Mauri gives us a quick look at his pre-con, Agile Data Warehousing: Start to Finish.

Q: If your pre-con had a theme song, what would it be and why?
“One Vision” by Queen. The data warehouse is all about having one, coherent vision of all company data and information.

Q: What interests you most about the Agile approach to data warehouse design, testing, and implementation?
It gives results quickly, and everyone involved in it, from developers to managers, get excited because they start to understand the strategic importance of the data warehouse. As a result of being able to quickly get insights, they start to use it to democratize the information in the company, really changing how decisions are made and ultimately making the process much more efficient and based on “real” data, not just intuition or, worse, suppositions. This is the era of data-driven decisions, and the data warehouse is one of the backbones of that.

Q: What's the most surprising statement attendees might hear you say during the pre-con?
That the only thing that will remain constant is that requirements will always change. I’m quite sure that many will also be surprised to learn that there is no product or model or tool that can make you Agile. Agile is a mindset, and thus the most important thing is to start to think that way. Everything will become natural after that first step.

Q: What’s the biggest myth around the agile process that you’d like to debunk?
I have two: The first myth is that the data warehouse is now obsolete, and with Big Data tools like Hadoop and In-Memory technologies, one can live without it. It’s really an important myth to debunk, because a data warehouse is not only a physical container, but is the place where people expect to find correct data. It’s a sort of metaphysical concept that alone can change how decisions are made within a company. 

The second myth is about tools that enable you to become Agile. As I noted earlier, this is another myth I’ll try to debunk.

Q: What still excites you in the real world when using Agile data warehousing principles?
I love to see the huge impact a good BI/DWH solution has in a company. And I love even more to see the faces of managers used to waiting for weeks or months to have a new business requirement put into production when I tell them, “OK, you’ll have it in a couple of days.” It’s like being a superhero… except they don’t allow me to go to their offices wearing a blue and red spandex suit. :)

Q: If attendees could start putting into practice just one thing after this pre-con, what would you want that to be?
Automation. I’ll show attendees how they can automate up to 66% of data warehouse creation. This really changes the game because it allows you to support frequent changes and move in the direction of becoming Agile.

Check out our other Q&As with PASS Summit 2014 pre-con speakers.


PASS Summit 2014 Pre-Con Preview: Adam Jorgensen & John Welch

Oct. 7, 2014 – Go inside PASS Summit 2014's full-day pre-conference sessions in this Q&A series with our presenters. In this preview, Adam Jorgensen and John Welch take us inside their pre-con, Big Data: Deploy, Design, and Manage Like a Pro.

Q: If your pre-con had a theme song, what would it be and why?
Adam and John: "
The Imperial March" (Darth Vader's Theme) – Big Data is going to take over the world.

Q: Adam, what excites you most about helping attendees unlock the potential of Big Data?
I want to help them understand that they already work with data, big and small, but now they have tools to help them be more agile and proactive in their design and approach to this analysis.

Q: John, what's the most surprising statement attendees might hear you say during the pre-con?
There's no such thing as unstructured data. A lot of people associate Big Data with unstructured data, but really, all data has some sort of structure. If it didn't, it would be useless. It's more a matter of when and how the structure is applied to the data.
 
Q: Back to you Adam: What’s the biggest Big Data myth that you’d like to debunk?
That it has to be "Big" – size doesn't matter‏. The tools work regardless of the data size and offer new ways to work with any size of data.

Q: John, what still trips you up in the real world when designing, deploying, and managing Big Data solutions?
Staying on top of the options available. Relational databases are relatively stable at this point, in terms of the fundamental technology and approaches. Given a specific problem, there are generally known, proven approaches to solving it. Big Data is still rapidly evolving, and both the technologies available and the approaches are changing as people try to solve new problems, or solve older problems better. 

Q: If attendees could start putting into practice just one thing after this pre-con, what would you both want that to be?
Adam and John: That they can start working with these technologies today, with minimal investment, and leverage existing skills while learning new ones.

Check out our other Q&As with PASS Summit 2014 pre-con speakers.


PASS Summit 2014 Pre-Con Preview: Kalen Delaney

Oct. 7, 2014 – Go inside PASS Summit 2014's full-day pre-conference sessions in this Q&A series with our presenters. In this preview, Kalen Delaney gives us a sneak peek at what to expect from her pre-con, In-Memory OLTP Internals: How Is a 30x Performance Boost Possible?

Q: If your pre-con had a theme song, what would it be and why?
“It’s a Whole New World” (music by Alan Menken and lyrics by Tim Rice) because it IS a WHOLE new world – a completely new way to think about data and data management. And it’s almost like magic. :)

Q: What excites you most about In-Memory OLTP?
The most exciting thing is the potential for orders of magnitude performance increases with a very gentle learning curve.

Q: What's the most surprising statement attendees might hear you say during the pre-con?
If I told you, then it wouldn’t be a surprise!

Q: What’s the biggest myth around In-Memory OLTP that you’d like to debunk?
The biggest myth is that In-Memory OLTP is not really all that different from what we’ve had before or that it’s just a new implementation of the old DBCC PINTABLE.

Q: What still trips you up in the real world when working with concurrency issues and trying to implement the fastest SQL Server solutions possible?
The biggest thing that trips me up is thinking people really understand the dangers of using NOLOCK when they don’t. They use it anyway! So with In-Memory OLTP, they’ll never even have the inclination to use NOLOCK.

Q: If attendees could start putting into practice just one thing after this pre-con, what would you want that to be?
I hope that people will decide to give In-Memory OLTP a try and, in a test environment, start converting some of their performance-critical tables to memory-optimized tables to see how easy it is and how much difference just a tiny change can make. 

Check out our other Q&As with PASS Summit 2014 pre-con speakers.


PASS Summit 2014 Pre-Con Preview: Marco Russo

Oct. 6, 2014 – Go inside PASS Summit 2014's full-day pre-conference sessions in this Q&A series with our presenters. In this preview, Marco Russo shares some of what you’ll learn in his pre-con, Data Modeling in SSAS Tabular.

Q: What excites you most about the SSAS Tabular model?
It is simple to learn how to create an SSAS Tabular model, and is very fast and powerful at the same time. You can achieve great performance goals on a large volume of data, even with complex calculations involved. 
 
Q: What's the most surprising statement attendees might hear you say during the pre-con?
Forget the traditional way of modeling data, worrying about the size of each data type. Now you have to think about data distribution.

Q: What’s the biggest DAX myth that you’d like to debunk?
Even if we won't talk too much about DAX (there is DAX-specific pre-con on Tuesday), I will quickly show that DAX is more similar to SQL than to MDX, whereas some BI developers still think the opposite is true.

Q: What still trips you up in the real world when creating and implementing Tabular data models?
The lack of a great DAX Editor!

Q: If attendees could start putting into practice just one thing after this pre-con, what would you want that to be?
I'd like them to quickly map their relational model into a Tabular model; they can get their first complete Tabular model up and working in less than 1 hour.

Check out our other Q&As with PASS Summit 2014 pre-con speakers.


PASS Summit 2014 Pre-Con Preview: Paul Randal

Oct. 6, 2014 – Go inside PASS Summit 2014's full-day pre-conference sessions in this Q&A series with our presenters. In this preview, Paul Randal shares some quick tidbits from his pre-con, Performance Troubleshooting Using Waits and Latches.

Q: What excites you most about using wait and latch statistics to investigate performance problems?
It provides one of the easiest ways to get an idea of what's happening with a poorly performing workload and can help avoid a lot of wasted time when performance troubleshooting.

Q: What's the most surprising statement attendees might hear you say during the pre-con?
It depends. :-)
 
Q: What’s the biggest myth around waits and latches that you’d like to debunk?
Probably the worst myth out there is that if you see a lot of CXPACKET waits, you should set your instance MAXDOP to 1.

Q: What are a few of your favorite wait and latch types?
SOS_SCHEDULER_YIELD, PAGELATCH_UP, and WRITELOG.

Q: If attendees could start putting into practice just one thing after this pre-con, what would you want that to be?
Using sys.dm_os_waiting_tasks to immediately see what's happening on a SQL Server instance.

Check out our other Q&As with PASS Summit 2014 pre-con speakers.

PASS Summit 2014 Pre-Con Preview: Erin Stellato & Jonathan Kehayias

Oct. 6, 2014 – Go inside PASS Summit 2014's full-day pre-conference sessions in this Q&A series with our presenters. In this preview, Erin Stellato and Jonathan Kehayias give you a sneak peek at Everything You Never Wanted to Know about Extended Events.

Q: If your pre-con had a theme song, what would it be and why?
Jonathan:
“Let It Go” from Frozen, because we are all going to eventually have to let SQL Trace go. I'm surprised I even thought of that, I'm generally not someone who would think about a theme song.
Erin: Jon and I both have daughters, one year apart in age – his answer is perfect. :)

Q: Erin, what excites you most about working with Extended Events?
The flexibility of it. You have the ability to do so much more – compared to Trace/Profiler – when capturing data that it gives you better insight when you're troubleshooting a problem or just trying to understand what's going on inside SQL Server.   

 Q: Jonathan, what's the most surprising statement attendees might hear you say during the pre-con?
I still use Profiler and Trace some times. Old habits die hard, even after writing 400+ pages of content on Extended Events since 2008.
 
 Q: Back to you, Erin: What’s the biggest myth around using Extended Events that you’d like to debunk?
That Extended Events are difficult to understand and figure out how to use. Before the UI in 2012, I admit that they were not intuitive. But between the UI and the approach we take when teaching, I think people will see how easy XE is to grasp and immediately start using.

Q: Jonathan, what’s your favorite advanced XE troubleshooting method that isn’t possible with Trace?
I would say collecting a memory dump and callstack information for a specific error or event occurring one time, and one time only, to minimize impact to the system over previous methods.
 
Q: If attendees could start putting into practice just one thing after this pre-con, what would you both want that to be?
Replacing old methods of collecting data using Trace and Profiler with a new understanding of Extended Events and the UI in SSMS.

Check out our other Q&As with PASS Summit 2014 pre-con speakers.