PASS Summit Survey Results


The good news is that the people have spoken.

The bad news is there were less than 200 people involved.  I’m hesitant to read too much into a survey with such a small sampling of a big audience, but hey, it’s what we’ve got.  The PASS 2010 Summit survey results are out, and a few things struck me as interesting in the results.

Question 4 asks if they’re going to attend the 2010 Summit in Seattle, and I’ve simplified the answers a little by lumping them together:

I’m cheating a little there because I lumped some answers together.  Roughly 16% of the survey responses are from people who are not attending.

If you want to increase attendance, you can’t just focus on the people who are ready to hand over the check.  You want to examine the responses differently.  You have to satisfy the people who are ready to fork over their cash, because you don’t want to lose their business, but you have to listen differently to the people who tell you they’re not coming.  See, these folks were willing to take the time (a LOT of time, I might add, because this survey was ridiculously long) to tell you what they’re looking for in a Summit.  This is why it takes a long time and a lot of labor to analyze results like this, and why I admire the PASS volunteers for putting so much time into listening to the community.

Question 7 asks who you’re very interested in hearing speak, and I’ve reorganized the bejeezus out of this answer.  The question asked how interested you were to hear each type of speaker, and I only took the scores from people who marked “Very Interested,” not “Somewhat Interested” or below.  I want the kinds of sessions people really get excited about, and the results were:

A whopping 92% said they were very interested in real world speakers.  Second place went to “Established Industry Speakers,” which I would take to mean trainers and authors.  Less than half of the responses had such an excited response to any Microsoft presenters.

And who are these “real world speakers”?  They’re YOU.  If you’ve been thinking nobody wants to hear you speak at PASS, you’re wrong.  Now would be an excellent time to register for Chuck Heinzelman’s presentation on creating a winning abstract.  It’s completely free, and 92% of your peers are very interested in hearing what you have to say.

By the way, the next time PASS tells you how desperate people are to hear Microsoft presenters, remind them of the magic number: 92% real world. Community is number one.  We’re here for each other first, and everybody else comes second.  Further evidence can be found in question 12, which asks why people come to the Summit.  Access to Microsoft employees ranked lower than education, networking, and access to experts.

Question 2 asks “What job description best describes your main job responsibilities?” With that kind of wording, you might expect to be only able to pick a single answer, but no.  Attendees were allowed to pick multiple responses, so they add up to more than 100%.  This is what happens when you let database people design the user interface, ha ha ho ho…put the stick down, people.

The majority of attendees are DBAs and developers, and if you include architects, that’s almost 75% of the responses.  Unfortunately, the armchair statistician in me wants to throw out this answer because I can imagine there’s people out there who marked both DBA and developer, whereas no architect would ever check more than one box for that question, because they understand how this question was supposed to work.  I kid.  Architects don’t understand that stuff either.

Anyway, my point is that with such a majority of DBA/developer types, you might draw the conclusion that the majority of the training should be focused on the SQL Server engine (not BI).

Question 39 would seem to answer that, but the question wording doesn’t support decisionmaking.  Question 39 asks if you want to give your opinion about BI topics, and 43% said no.  You can’t take that to mean 43% of the attendees don’t want to attend BI presentations, though, because for each BI topic, some people voted that no, they weren’t interested in that topic too.  The only way to interpret this is to say that at least 43% of attendees don’t want to hear about any BI whatsoever, and above that, some percentage of people aren’t interested in various BI topics.  You can’t say that 57% of attendees want to hear about BI.

Question 21 asks what subtracks you’d like to see, and I gotta confess that I’m a little surprised by some of the answers:

  • Performance Tuning – 78% very interested, 2% say no.  Wow.  Looks like I’ll be submitting tuning sessions this year.
  • DBA 101 – 26% very interested, 38% say no.  I always wonder what the mix of experience is.  26% sounds low, but turn it into a ratio, and it means that 1 of every 4 attendees would like to see basic-level courses.
  • SharePoint – 22% very interested, 41% say no.  I would have guessed this number would have been a lot higher given the SQL 2008 R2 release with PowerPivot and all the sexy BI stuff built with SharePoint.

Question 24 asks what DBA sessions you’d like to see, and one of the answers stunned me.  29% of DBA responses said they were very interested in a session on Backup Compression.

Are you kidding me?

You check the box and it goes.  End of story.  How the hell do you make a one hour session out of this?!?  Are you people just checking boxes because you like the SQL feature or something?  Mark my words, somebody’s going to do at least one session on it, and they’re not going to get good speaker feedback because the topic is so damn boring.  As a speaker, you gotta do a good job of picking topics.  Just because people want to hear about something doesn’t mean you should do a presentation on that, because you’re shooting yourself in the foot.

Even funnier, Backup Compression outranked Transparent Data Encryption, Filestream, and Spatial Data, all of which require more forethought and planning than compression.  If I was the product manager for those last three features, I’d be dunking my head in the toilet right now, because that’s just depressing.

Question 16 asks if PASS should offer longer sessions, but less of them with more details:

As a presenter, I’d love to see a few deeper-dive sessions.  I wouldn’t make all of them longer, but just say one long session per day – perhaps the first session after lunch.  People are moving slow then anyway, and it takes them more time to get up to speed.

I’d also like to see “lightning talks” – a 30-minute block at the beginning of the day, right after the keynote, where 6 presenters talk rapid-fire for 5 minutes each.  Let them “sell” attendees on why their presentation is so good.  And maybe lop off some time on the keynotes, because those haven’t moved nearly fast enough.  (Especially the all-marketing ones – dear Lord, those are bad.)

There’s a lot of interesting stuff in that survey, and if you’re thinking about submitting abstracts for the PASS Summit, you should dig through there.  Keep an open mind about the stuff you know versus the stuff you think people want to hear.  I never would have submitted a social media session last year if Jason Massie hadn’t suggested it, but we ended up in the top 5 ratings for the Professional Development track.  Now, looking at the survey results, I think I’ll submit another one this year, but this time I’ll focus on marketing yourself and technical presentations.

Side note – want to stick charts like this in your blog posts no matter what blog software you use?  Check out, a slick little GUI front end for Google Charts.  Google Charts takes a simple URL and turns it into a chart.  So when I put this HTML code into my blog post:

<img src=”×200&amp;cht=p3&amp;chtt=Press Your Luck&amp;chd=s:9Z&amp;chco=bbbbbb&amp;chxl=0:|Big Money|No Whammies|&amp;chxt=x”/>

You see this:

Spiffy, free, and works anywhere.  Enjoy!

Previous Post
#SQLR2 Launch Goofs
Next Post
How to Get Readers to Pay Attention

54 Comments. Leave new

  • On Question 7, just FWIW, my opinion about the importance of Microsoft presence at PASS Summit is *NOT AT ALL* about Microsoft *presenters.* It is about having gobs of people from the dev team roaming around, taking part in casual conversations, and participating in side meetings for our specific group. How often do you see UC or Bill or Boris or Tobias or Isaac or Michael in Charlotte or Dallas or Chicago? I value that kind of opportunity highly, since I know that there will be quality speakers and valuable sessions no matter where PASS is held. And it’s no question that the attendance of those folks would drop off if PASS were held elsewhere.

    • Aaron – well, there’s two questions. Question 7 specifically asked, “Please indicate how interested you are in attending educational sessions based on the speaker type,” and that’s where attendees spoke loud and clear about community over Microsoft.

      Question 12 asked “How important to you are the following reasons to attend the PASS Community Summit?” and had answers about access to Microsoft staff. About 1 in 3 responses said that Microsoft staff were “very important” to their Summit experience.

    • Aaron,

      I said this on a previous post somewhere, but I don’t think most attendees care about the MS people.

      Every time I walk into the areas w/ MS people, it’s almost 100% MS people in there. A few years ago I spent almost three hours talking with a few guys from the SQLCLR team, and NO ONE else was around for the entire time. Great for me, not so great for attendees in general.

      Why is this the case? Well, it could be because the MS people aren’t, by and large, very friendly. They sit at tables and chat with each other and don’t make any effort to engage. Best case scenario, you’re walking into a conversation that you need to interrupt. And most people won’t do that. Worst case scenario, they treat you like you’re a maggot. Last year I tried to approach an ADO.NET PM who I didn’t know and the guy totally brushed me off. This is absolutely not the case with everyone (e.g. UC), but a lot of them are there simply because they’ve been told to be there, not because they actually want to interact.

      The other reason? It could be that attendees simply don’t want to talk to MS people. Not sure what the truth is there.

      • Adam, I think you’ve really hit on something here. Last year, when I approached the SQLCAT team to talk about setting up TFS, I was given the “what are you, a moron?” attitude by them. Trying to use the resources made available to me, I was hoping to get more clear information about a good setup of the environment and maybe some tips and tricks to make it super-cool. Nope. Instead, I got the brush off and left the table after not even being there 5 minutes. Sad. Truly.

        • When we run customer events at Quest, we have an informal rule: no more than 2 Quest employees are allowed to stand together, and Quest employees are not allowed to outnumber customers. When there’s too many vendor people and not enough customers, the customer feels outgunned and intimidated, and it creates bad interactions. The Microsoft areas at PASS have always felt intimidating to me – it felt that way before I came to work for Quest, and it still feels that way to me today.

          • I wonder if we should ask the PASS leadership to limit the ‘softies to just a few – to cut down on this sort of thing. It’s supposed to be a benefit, but if it isn’t, let’s stop sending folks out there, by all means? I’ll bring this up on the Microsoft side and see what we can do.

          • I’m setting up a call on our side this morning to talk about this. I hate to see what should be a good thing go wrong – let’s see what we can do to make things better.

        • The results were eye opening to me. Especially with CSS towards the bottom. Last year I thought that the SQL Clinic room was the busiest it had every been, and it was a joint effort between CSS and SQL CAT, although SQL CAT outnumbered CSS by a lot. There were times where I was the only one in the room from the CSS side.

          Don’t know how the SQL Clinic room falls into that. Although, I’m sure we can definately improve based on some of the comments above. But this was seperate from the Ask the Experts area.

          • Adam: I felt that the SQL Clinic room had a much friendlier “vibe” than the ATE area. And a big help was the person manning the desk at the front and helping guide people to the right person to talk to. The ATE area could probably be much improved by the addition of someone–or a few people–in that role.

      • Adam,

        I also bounced off of a few MS people last year in the MS section and was very surprised by it. I honestly expected the people that would be there would be the people interesting in helping. However, I also found several MS people willing to help, so I think it’s split somewhat.

        But, yeah, I was surprised by how few attendees were there. I kept expecting to have to push my way through in order to shout a question over the crowd, but instead strolled in and started a conversation (except when the MS people were nasty).

      • Adam,

        I asked the SQL CAT and SQL Dev team a question last Summit and got the run around on who would know the answer. In the end, they didn’t know and we’re telling me that I was going to need to open up a case. I asked you the same question and you gave me what turned out to be the correct answer right away.

        Why should I bother talking to them??

  • Sure Brent, but isn’t it possible that a lot of the answers to question 12 were swayed by the greater knowledge that if they say they *are* interested in access to Microsoft staff, the Summit will continue to be held in Seattle?

    The problem with the survey is that an overriding agenda can greatly influence almost every answer.

    • Yep, this is why you get survey professionals to build surveys, and database professionals to build databases, and not the other way around. You have to strip all of the bias out of the survey, or else the numbers are meaningless. That’s why I didn’t even bother responding to question 12 – I think it’s not good for making decisions.

  • Love the post, Brent! Yes, survey results often require lots of interpretation and can easily be skewed in many cases. My personal interpretation would be in line with yours, though. As for the more than 100% totals on questions, you are dead-on! As I read through those, I had to ask myself “Well, what spin are they going to take on those?” I hope it’s more along the lines of how you presented it here.
    One thing, though – it’s really difficult to tell those vendor kenotes “sorry, you’re just boring and not popular” because….wait for it….wait….those are HUGE revenue generators for sponsorships! It’s a sad truth, but money we need to make the event successful.

  • Speaking of statistics and their misuse, would it be possible for a Microsoft person to be an “Industry Expert”? Perhaps when the question was asked the respondent didn’t see a complete distinction, which might skew the answsers.

    Some of us still actively work as a DBA – and since I get to see DBA shops from large to small each week, would I have a perspective that might be interesting to share? Just thinking out loud here – no offense taken or implied…

    And don’t take this as a “we should stay in Seattle” thing – I’m ALL for having the event in Florida or Hawaii or some other beautiful spot. I couldn’t go, of course, but I would still love to see it there!

    • According to the survey design, no. Those answers were pretty bad choices. My first pass at a set of choices would be:
      – Practitioners – people who spend their time working out in the field on the same task for months with the product either as employees or consultants
      – Teachers – people who make a living teaching other people how to do tasks or writing books
      – Support – people who work at Microsoft or consulting companies, spending short bursts of time doing very advanced work with different clients each time
      – Developers – people who work at Microsoft writing the code inside the product

      Each of these groups of people has their own unique advantage – some are better at going deep with a feature, some know how it works in real life, and some are better at communicating ideas in front of crowds. There’s people from Microsoft in all four of those categories, and there’s people in the field in three of the four.

    • How can you be an industry expert when you’re high on that blue Kool Aid stuff they make you drink twice a week?

      Seriously, though, I think MS people are expected to be experts in their area already. “Industry expert” means a third-party expert, who can (hopefully) provide a more balanced point of view, with no fear of losing a job, etc.

      • I don’t fear losing my job – I think everyone knows that!

        And by “balanced”, you mean that we aren’t? That’s a little harsh.

        • Buck,

          Yes, I do mean that “you” aren’t. Although “you” (Buck), usually are, “you” (Microsoft people in general) are quite often not AT ALL. Quite a few people on the product team suffer from a serious case of head-in-the-sand syndrome and will absolutely not admit that there are design flaws in new product features, etc.

          • Wow. Brutal. Sorry to hear that. I admit mistakes all the time – check any of my posts/writings, etc. Shame to hear that you feel that way.

            Why be an MVP for a product like that?

          • Seriously, Buck? I know you’re well aware of the people I’m referring to, so let’s not play a game here. The very fact that you’re pretending like there is no issue proves my point.

            Why am I an MVP? MVPs are “independent experts”. I love working with the product, I love teaching people how to get more mileage out of it, and I absolutely do not, and will not (at least, not until some day if I take a job with MS), evangelize for the sake of selling licenses. That’s Microsoft’s job.

          • Not playing – just truly sorry to hear that you’ve had these issues. And I’m sure there are people who have to toe a line – I just won’t do that. If I have to say something I don’t believe in to keep a job, I’d quit. Most of the folks I know are the same.

            That doesn’t mean that you can’t look at the positive side of an issue, or a negative side. As long as we’re talking about someone truly ignoring your issues, then I’m with you. Bad dev. Trust me, I nail people on this every day!

            Thanks for the passion, my friend. Keep it up!

          • Buck asked a really good question here – “Why be an MVP for a product like that?”

            I get involved with the SQL Server community for two reasons: SQL Server and the community. It might sound obvious, but hang in here for a second. First, I really do like the database server software itself. It empowers me to make a great living doing great stuff. The more I know about it, the more I can do with less effort, and the more money I make. Second, I like a lot of the other people who are using SQL Server. There’s a lot of funny, smart, enjoyable people at PASS.

            But I don’t get involved with the SQL Server community to interact more with Microsoft.

            Don’t get me wrong – there’s a lot of flat-out AWESOME people who get paid by Microsoft, and they happen to like hanging out in the community too. However, I get the feeling these people (like Buck and Jimmy May) would be a part of the community no matter who they worked for.

  • Nice blog entry Brent, however I have to disagree with you on question 24. Backup compression doesn’t always work! In my environment we use Transparent Data Encryption and once that’s enabled native backup compression goes out the window. I even did the rounds on the vendor stands at the last SQLBits conference and neither Red Gate, Idera or Quest could guarantee significant compression with their products when TDE was turned on.

    So you could do a session on TDE with the slant on how it affects backup compression.

    • Richard – yep, anytime you have “random”-ish data, it won’t compress. Binary files and encrypted data are random, so they don’t compress. But still – there we go, we just said it, and we’re done. How else can we get an hour out of it?

      • Agree that it’s a “thin” topic, but I think it indicates interest rather than thinking it through about what a briefing like that would comprise. Were it me, I would title a briefing “Backup and Restore Strategies, including Compression Considerations” to draw interest.

        Darn. Just gave away my topic!

      • Regarding compression as a topic, I agree that it is probably stretching it to make a 60-minute session on that one topic, although you might get 15 minutes out of talking about the effects of the different options (Page vs. Row) and their effects on other things like backup compression. I have read a couple of interesting blog posts comparing space savings and so on. So to me that level of interest combined with the “thinness” of the subject says don’t try to make a full presentation on that one topic, but instead be sure to include a section on that topic in your larger presentation on Storage Considerations for the DBA or DBA 101 or Top 10 Options to Mess With.

        By the way, I full agree with your point that you should work with a survey professional to get a high quality, non-biased survey. It is damn tricky to write survey questions that do not reveal your own biases. And it is more than just finding mutually exclusive answers, although that by itself will go a long way toward building a better survey.

        • Remember – the answer was BACKUP compression, not compression. Row/page are options on data compression, not backup compression. Backup compression is literally on or off.

          • D’oh! You’re right. I forgot that important little fact when I got engrossed in some of the other comments. Still, if it’s that popular of a topic, it might be worth a mention in your abstract and 3 or 4 minutes of time to talk about it, even if it’s a story about how long you have waited for this feature and how much space it is saving you; or how you don’t care about this feature because you have used LiteSpeed for years.

            Give them what they want, and then give them even more (value) than they asked for.

  • By the way – another comment here. I’m distressed to hear that ANYONE is mistreated by a ‘softie. Please, please, please let someone else in blue know if that happens, and we’ll make sure that person NEVER comes to another event. It’s a great privilege for us to be able to go, and we can get into TONS of trouble for that behaviour. I take it very personally when that happens.

    If you do see a table of Mircosoft people “talking among themselves” don’t let that be a barrier! They have to have something to do – and in my experience I have never had a blue-short ignore anyone just because they were chatting. It’s just a way to pass the time until you come by and ask us something. All of this is volunteer, off of our own time, and taking us away from work. Most of the time we have to work a weekend to catch back up, but it’s totally worth it to chat with you.

    So pull up a chair!

  • Add me to the list of folks that blown off by a MS person last year. In years past it wasn’t this bad. Any of you who know me, I’m not afraid to walk up to anyone and start a conversation. To be completely honest I was blown off by Adam last year too.

    • Ouch! To be fair, I’ve blown people off when I’ve been hustling through the halls trying to get from one session to the next. I hope I’ve never blown anybody off after the sessions were over though.

    • Ouch, indeed! I apologize if that happened; I certainly didn’t mean to do that and have no recollection of doing so. Perhaps the MS people and I should all attend a training course on how to better interact with others

    • Wes, sometimes it IS timing. I walked into a group of about 6 people at the Welcome Reception, and seconds after saying “Hi” it was only me and Joe Webb standing there. Kind of embarassing, but I’m sure it wasn’t personal…just bad timing on my part.

    • However, I’m quite sure that was NOT the case at the SQLCAT area where it was just a bunch of MS people talking to each other.

      • SQLCAT did have a few good sessions last year, especially the HA roundtable which was totally customer driven.

        We’ve all blown people off at times – usually not on purpose. I know if I’m hustling to get to my room to set up and speak, I don’t have time to chit chat, but then again, I don’t ignore people. I’m polite about it.

        Remember kids that Buck is not your average MS person 🙂

  • Have to jump in after seeing the conversation on twitter.

    I’m also disappointed in the small number of responses. Perhaps a better time for a survey like this is just AFTER The Summit where you might drive up responses because the experience is fresh.

    As far as interaction with the Developers, I definitely think that is something that is more important to those who are heavily invested in the community like the bloggers and twitter people and an even smaller subset that are really pushing SQL Server. I’m in a small environment and there’s very little that I would need to ask a person on the SQL Server development team. If I have a question about something I much more likely to ask an industry expert, for example with Extended Events I’d ask Adam or Jonathan Kehayias not a person on the product development team because I want to know how and why to use it, not how SQL Server implements it. Now Adam or Jonathan may want to to know those details , but the other 90-955 of us don’t.

    For the record, the MS people I have interacted with, not many but some, have been helpful. Yes, most drink the Kool-AId, but they are polite and helpful.

  • It wasn’t that big a deal to me. I know Adam is in demand. My boss standing next to me was a little flustered by it.

    I didn’t take it personally.

    Wendy, it could have been a timing issue.

    I’ll try again next year 🙂

    • Wes, I know it’s no big deal at this point, but could you e-mail me and remind me of the situation? I just want to relive it for a moment 🙂

      adam [at] sqlblog [dot] com

      … and tell your boss I apologize.

  • Actually, questions 8, 9, and 10 are pretty interesting to me – the 52% where less than half (probably meaning all) of their deployments are SQL Server 2005 or earlier, and 45% or more of those are 2005, with a whopping 34% at 2000 or earlier. Wow! For all of us that focus on 2008/2008 R2 content, PASS still needs to help those stuck at earlier versions. Like it or not, 2000 and 2005 are here to stay in some orgs. T

    I still need to go through the whole thing in more depth, although the good news is that there is still interest in HA 🙂

  • I’ve only been to the one summit, a couple of years ago, but found the MS Dev and SQLCAT folks to be really approachable. I think I actually stood for over 90 minutes with one of them discussing filestream data types. I think having them there and available is very important even though they may not be the same folks that you would want presenting.

    As regards a session on data compression, that more than likely comes back to the 101 type session. While it’s very simple to enable there is a general lack of awareness that you can do so with a quick check of a box or sp_configure change. People also like to know what it does and how it does that compression. Do we really NEED to know the internals difference with ROW and PAGE compression? Probably not, we just need to know the impact when they are enabled, however that does not stop us from finding out as much as possible before implementing, or trying to grok how it actually functions and why one has more impact than the other.

    It’s a shame that there were so few respondents overall as the numbers could easily not be truly representative of the community at large.

  • Well, I didn’t accept an FTE position at Microsoft until last year. So does that mean I can no longer be considered an industry expert? Not that I’m claiming to have ever been referred to as one.

    Personally, I’d like to get a little insight into how MSFT decides who to send to the Summit as the CSS experts. I work in CSS and I can tell you that there’s no call out for people who want to represent CSS. I have no way to determine how they make that decision.

    It seem to be that when they talk about experts from CSS, what they really mean are the PFE’s (Premier Field Engineers) and nobody else.

  • By the way, Brent, interesting notes on the “backup compression” answer. I found the survey itself to be very strangely worded with regard to topics. CHANGETABLE — a function used by CDC — was listed as a topic of choice. If I knew nothing about CDC, I might check that box, because maybe I’m interested in finding out what it is. But no one could possibly do a one-hour session on JUST that function.

    Most people who responded probably don’t know that backup compression is either on or not, so they checked the box.

    The problem here is a survey that was generated based on what I can only assume is some list of buzzwords supplied by Microsoft, rather than real-world categories.

  • Paul White NZ
    April 26, 2010 11:11 am

    On the question of Backup Compression – is it possible that people were expressing an interest in the new compression features as a whole, not just Backup Compression?

    The relationships between Data Compression, Backup Compression, and things like TDE are fascinating.

    I could easily do an hours on the content of for example :c)

    It’s good to see backup compression made it into R2 Standard Edition, shame about data compression.

    • Paul – that’s the problem with trying to read your own opinion into survey results. You’re explaining something that’s not on the survey. They answered “backup compression” – not compression, not data compression, but “backup compression.” I agree that data compression and TDE are interesting, but that’s not what the survey asked.

      • Paul White NZ
        April 26, 2010 11:22 am

        OK – I haven’t seen the survey so was just guessing. Many DBAs seem confused by the differences, so I was just wondering if there might be more to it.

  • Great writeup, Brent, and I mostly agree. I’ll need to respond and drop some thoughts in a blog to trackback to here.

  • Brent,

    To comment on question 21 as to what topic I’d like to hear, it comes down to what do I use everyday. Performance tuning is something DBAs deal with frequently it can make a big deal in your user environment. SQL 2008 R2 release with PowerPivot and all the sexy BI stuff built with SharePoint is off in the future for me; I probably won’t see it for awhile. I’m much more interested in what can help me here and now, not in the future.
    I hope you do a few sessions on performance tuning, sign me up!

  • I personally do not like most performance tuning presentations I’ve seen so far at MS SQL Server events, for lots of reasons. Many methods are lacking, in my opinion. It is a great topic to pursue, however.

    I’ve been to many SQL Saturdays, viewed PASS videos, and read a lot of books. I’d obviously learn more in some focused training events. But, I have seen many excellent presentations. In Grant Fritchey’s PASS presentation he matter-of-fact answered one question I’ve had and asked, but never got a clear answer until his.

    I prefer presentations that focus on one specific strategy, e.g. user response time, managing server bottlenecks, etc. I think K.D. covers the general strategies in her 2k5 performance tuning book.

    I have yet to see a good presentation on end-to-end application response time tuning, where we proportionally measure and show where the user’s business task waits from SQL Server perspective. I’m talking beyond SQL statement tuning, e.g. to measure the waits events like network latency, cpu, io, locks, blocks, etc. I’m interested in wherever the time is going, proportionally.

    Also, I’d like to see more presentations of index evaluation and comparison used by query groups, rather than looking for one SQL statement’s missing index. Lets assume and discuss a list of queries and evaluate the set of indexes they used. Make sense?


    • Hi, Mike. Those are great questions, but those are way, way outside of the scope that someone can cover in an hour-long session. This is why I launched this week – I hear these kinds of requests from senior-level DBAs pretty frequently, and I wanted a venue where I could dive deeply into the same topic for 3-4 hours straight.

      A challenge of end-to-end application response time measurement is that different client applications demand such different tuning strategies. For example, I was recently working with a company whose web tier was composed entirely of Java VMs in Linux. Measuring the end response time involved using web testing software, virtualization management software, and more. The strategy we used there is completely different than the client I was at this weekend, who did analytics with a thick client front end.

      It sounds like you’re ready to move up past the free events and graduate to more senior ones like pre-conference and post-conference events that really dive deeply into a subject instead of doing a lot of one-hour wading.


Leave a Reply

Your email address will not be published. Required fields are marked *

Fill out this field
Fill out this field
Please enter a valid email address.