My name is Brent Ozar, and I help people make SQL Server go faster.
That basically means I measure things for a living.
If I’m going to make something faster, I have to start by knowing exactly how slow it is.
When I first started working with SQL Server, I just measured it with Perfmon metrics. I spent so much time capturing metrics, interpreting what they meant, documenting them, setting up email alerts about them. Then I found out about DMVs – SQL Server’s dynamic management views. I started querying those, and wow, there was so much information available about internals, what was running now, what had run recently, how healthy the server was, you name it.
There was so much information available that I couldn’t possibly collect it all – so I bought monitoring software that captured everything. It would analyze and trend it all for me, give me nice big dashboards, and it would even send me alert emails. Page life expectancy is too low! Disk queue length is too high!
I got so many monitoring alerts that I had to set up a rule in my email to push all of those alerts into a folder. And then I didn’t even read the alerts anymore. Sound familiar?
We’re all addicted to gathering trivial numbers.
Companies do this too – not just with their SQL Server data, but with all kinds of data. This is where the “big data” trend is coming from. It’s so cheap and easy to gather all kinds of data, but nobody knows what’s important, or how they should act on it. They just collect it all and stash it away somewhere, hoping that it will be useful someday. I’m not saying big data is bad – it’s awesome, when you have a use for it. But so often, it’s just like all those Perfmon counters and DMV metrics that we gather – it just lies around unused.
We even do this with our own personal lives.
I’ve got an Apple Watch on. (It’s no good, don’t buy one. Wait for the next version.) But one of the things it does is track the number of steps I take every day. I think we’re supposed to get like five thousand steps per day to be healthy. The first few days, I looked at it every day to see if I was active enough. I made sure I was taking five thousand steps, but, uh, I’m still fat.
That data point doesn’t matter for me. I’m just piling up more data that isn’t really relevant to my quality of life. What I need is for the Apple Watch to count the number of times I lift a beer glass. That would be a much more useful metric, right?
Finding the right measurement is hard.
I used to measure my career with the wrong metric.
I’m going to ask you to think for a few seconds about one number that you might monitor that tells how your job or your career is doing.
Was it money? Your annual salary?
That’s what mine was for a really long time. It’s how I got into databases. I always looked around the company for the most expensive thing. I wanted to stand near that thing, and fix it when it broke. When I worked in hotels, it looked like the computer was the most expensive thing, so I stood by that, and when it broke, I offered to help. Then I went to work for a hotel management company with lots of desktop computers, and I looked around again. The servers were the most expensive thing in that office, so I stood near those, and when they broke, I offered to help. I got so much server experience that I went to work for a software company, and I looked around for the most expensive thing there – it was SQL Server. So I read the SQL Server manuals and stood near that, waiting for it to break.
Along the way, I learned two things:
- If you fix expensive things, you make lots of money.
- Expensive things seem to break when I stand near them.
But because I used the wrong metric, my career went the wrong way.
I got a job working for a consulting company who said me: “This client is paying us to build them a data warehouse. The project budget includes a DBA, and we don’t have an extra one, so we’re going to hire you.” The money was fantastic – the most money I’d made so far.
But after I started, they told me a little something more: “We don’t actually have any work for you to do, so we just need you to sit in a cubicle, look busy, and don’t touch anything. As long as you can do that, we’ll be able to bill hours for your time, and you’ll make great money.”
That sounds awesome, right? Sit around and get paid!
But I wasn’t doing things I was proud of.
I wasn’t really helping the business, I wasn’t learning, and I wasn’t growing as a person. I was just putting in time for money.
My family life suffered. My health suffered. Even my dog suffered. My wife, Erika, would pick me up after work every day, and as usual, she had our dog, Ernie, with her. We started driving home, and on the way, Erika and I were both so stressed-out and tensed up that Ernie leaned over between the car seats and vomited. I’ve never seen so much vomit. I’ve been around drunk adults that vomited less than Ernie did that day. It was like you turned on a faucet and she just kept going and going.
Eventually the stress got to me and I tried something different.
I let go of my money measurement just a little bit, and took a lower-paying job that sounded really interesting. I went to work as a DBA with a team of fun people I really admired. I said to myself, I’d be willing to make less money if I could make a difference, and be more proud of what I did.
Those next few years were some of the best years of my life. I worked much, much harder, and I was making less, but I loved it. I learned virtualization, learned to manage shared storage, went to the PASS Summit for the first time, even started running.
I was doing things I was proud of.
So I started measuring my job by the number of things I was proud of.
I’m not talking about avoiding dessert, I’m talking about big things I want to tell my family about, just like a couple of things per year. Big things.
When you’re proud of something, you’ll work really hard at it. You’ll be excited by it, consumed by it, and you’ll get really good at it. And when you’re proud of it, and you’re good at it, it doesn’t even feel like work at all.
And it sounds odd, but you will make money at it.
You will make a lot of money at it.
This is the age of sponsored blogs, webcasts, podcasts, training classes, independent consulting – heck, even Twitch. If you’re good at playing games and you put a lot of love into what you do, you can even get paid for that! Granted, the competition is a lot tougher – but these are new careers our parents never dreamed of.
Here’s how to start measuring your job.
- Read Steve Kamb’s Epic Quest of Awesome. It’s just a web page. If you like that, read his book, Level Up Your Life.
- Read my version, my Epic Life Quest.
- Make a list of things you’d be really proud to achieve in the next couple of years. They can be totally big tasks that you have no idea where to start – that’s fine.
Today, at SQLSaturday, look at the session agenda, and here’s how you pick sessions:
- Pick sessions that get you closer to doing something you’d be really proud of.
- Pick sessions by presenters who have done the task you’d be really proud of – whether it’s building a certain kind of app, getting a certain kind of job, or starting a certain kind of blog. Talk to those presenters after their session. They’re – we’re – here because we want to help you achieve great things.
- If you don’t know what you want to do, pick sessions that scare you, sessions way outside of your comfort zone. That’s how you learn new things you never would have considered, and that might be the career path that turns out to be right for you.
Thanks for listening to me this morning, and enjoy your sessions at SQLSaturday Israel!