I was lucky to find some time in @cloudyadvice's busy schedule in September, and got the chance to spend some time chatting about his life, career, and insights into the world of cloud.
An AWS Community Builder holding AWS certs including Solutions Architect – Professional and the Advanced Networking Speciality, Mike is also a member of the FinOps Foundation, and speaks from decades of experience in the DevOps space. Find more from Mike on his website, and here on the Ops Community.
There's a full transcription below the video, in case you can't watch it. Feel free to Ask Mike Anything in the comments section and he'll respond asynchronously.
My name's Mike Graff. I'm currently Infrastructure Architecture Director at Dolby Laboratories, which is headquartered in San Francisco. My preferred pronouns are he/him. I've been doing IT for 30+ years at this point.
We are a company that focuses on the art and science of entertainment when it comes to audio-visual, all that stuff. I've been at Dolby for an incredibly long time for an IT person: I've been at Dolby for twenty two years.
I think that being able to constantly work on something new is what has kept me at the same place for twenty two years. I think that's what people tend to move to other companies for, they get sick of doing what they're doing. I've been lucky enough at Dolby to be able to work on new stuff every couple of years.
I started at Dolby as a (what was back then) was called a Network Analyst, so I was doing like basic networking, Windows Administration, Exchange email administration, and that kind of stuff, and kind of worked through the whole cycle at Dolby as far as doing that work - and then Server Architecture and led the company's effort to do virtualization. When I started, we had no virtual servers at all, so I led that effort to go to virtualization. I did a lot of work around colo's. So I'd started to do colo work and then in 2015/2016 started to get into Cloud. So I did the whole journey of physical servers to virtualization to colocation, and now we're doing Cloud.
Prior to this cloud work for two and a half years my project was: we built a new headquarters in San Francisco, and I got to do that whole project from from soup to nuts. Just starting from nothing, building it out, designing all the cabling infrastructure, designing all the server rooms, all the network closets and the wireless network and the conference rooms. I never want to work in another conference room again in my life! I hate that work! It's so annoying because your customers are never happy, especially at a company like Dolby where everybody who walks in the room is a sound engineer. You can imagine what that's like. So that's been quite fun and kept me fresh in my career, basically, and I've been able to drive forward and learn new stuff continuously, which just I think what we all want.
There are a lot of funny stories, some which I can share and some which I can't obviously, but I think the thing that's really stood out for me is how the journey that I've taken Dolby has kind of mirrored the way the industry has gone. I think that the thing that's kept me at this place for so long is I've been able to lead the charge on moving us to these more modern architectures, if you will. When I started, I walked into the server room and there was no racks, it was all just like white boxes. Everything was running on basically PC clones, and tape drives were stacked on top of each other, and people were having to come in and manually rotate tapes. So that was one of the first things - I was like, "Oh hell no, I am not doing that! So let's get a tape library!" Back then that was like, "Let's not be sitting having someone come in on the weekend to move tapes around." And then when that got out of scale and our tape library was constantly breaking down, I got to move the company to (we were very early in adopting) disk-to-disk backups, which was a lifesaver because at one point we were having to come in basically every single weekend to fix our tape robot because it was just completely overwhelmed with the amount of backups we were doing. Working at Dolby, you can imagine there's quite a lot of data to be backed up so going from tape to disk was a huge win for us. And then the whole virtualization thing, that was like a godsend. If you think back to the days when that first started, it solved a lot of problems that people were having around provisioning and time to build servers, and all that stuff, so that really was quite a quite a change that I really enjoyed. Back in the day our local utility in the Bay Area, they would actually write you a check for virtualizing servers because you were saving power theoretically. We had this program where we virtualized X number of servers, and they actually handed us check and we brought that to our finance department and they're like "What are we supposed to do with this check? Can we just take it for us and go buy some beer?"
I graduated college with an English degree and at the time I thought I was going to be a professor of medieval literature. That was what I studied in college, it was basically Chaucer and stuff like that, and so I thought that was going to be my my career path. I finished up college, and looked at the landscape and realized, "People have been doing this medieval literature thing for over five hundred years. What new can I bring to the conversation that has not already been said?" And so that kind of left me without a real career path in mind.
My first job out of college was working in a music store selling musical instruments (you can imagine how excited my parents when I took that job after them paying for four years of university education). And so I did that for a while, then I switched working for a computer reseller doing customer service. Anybody who's ever done customer service, my heart goes out to you. That is one of the crappiest jobs you can have - people basically just yelling at you all the time. But I was actually decent at it, you know, at defusing things, and that got me into a manager role. And then the company I was working at they just happened to have an opening in the IT department. It was a systems analyst job, very entry-level and I'm like, "Oh yeah, that's I think I want to do that instead!" I'd always been interested in computers from an early age. I had some very old computers that I won't mention, because they'll just a date me more than I probably want to, but I was always into that and so I thought, "Ok, I'll try this out!"
Again, this is a situation where people didn't look at my degree, or my previous job history. It was more about my track record at that company, and looking at how I stepped in and took on roles that were foreign to me. So that got me into that job and I worked at that place that was a computer reseller here in the Bay Area, that used to sell Macs before Apple put all the all the Apple resellers out of business with their own stores. So I worked on that job and that was working on VAX. I don't know if you've ever heard of a VAX - this is a mini computer, kind of like Mainframe style computing. That was what the whole company ran on, so I was doing batch processing and writing batch scripts and stuff like that on that on that job, and then they merged with another reseller, and I got the opportunity to become a Windows admin. They sent me to training. They trained me on, back in the day it was the MCSE.
So I got my Microsoft certification and I was off and running doing that job, and in 2000 I started to look for a new new gig and that was a great time to be looking for a job. I managed to land the job at Dolby, and I've been there ever since. Moving from doing the Systems Administration thing to moving up in that job and then becoming a manager of the team.
That project was a once-in-a-career kind of project. In my opinion, you know you don't get that opportunity very often, and it starts with your manager. My director at the time showed a lot of confidence in me and put me on this project saying, "You know what? We're gonna take you off, we're gonna basically make this your full time job, we're going to take away all your direct reports, and you're just going to go back to being completely hands on, running this project." I had done a fair number of data center builds. Dolby in the old days had a bad habit of building data centers in our offices, and so I traveled all around the world building data centers in different Dolby offices. So I had worked that type of a project on a smaller scale and this is just a much larger-size effort.
I think the other thing is having people around you who are experts on things. I'm not an expert in cable troughs and fire blocks and all the things you need to do when you're running cables through a seventeen-storey building, but you bring in the right people who do and you make sure they know what they're doing - and then your job is to orchestrate all those folks. There are a lot of frustrations. Dealing with Facilities folks can often be frustrating. That's been probably the best thing about working in the Cloud is I no longer have to deal with Facilities people...
I think that success in IT, or in any career, is really beyond the tech. It's about the soft skills and understanding the business needs. That's a big part of being successful in my opinion.
I think there's a couple things that I have that that served me well. One of my strong suits is understanding and researching and adopting new things. One of the things I do well is I can look at these new technologies, play with them, figure out where they might fit in the landscape. I think that's really helps me a lot, and that ability that I've demonstrated over the years to my management has given them the confidence to trust me and put me in the role that I'm in now.
I think the other thing that I have that I always say is my secret weapon is, I don't have a computer science degree. My degree is in English, English Literature, so from very early on it allowed me to do something very well that most people in IT can't do, which is communicate, both spoken as well as written communications. I think that was my big differentiator early on in my career, compared to my colleagues, and what allowed me to move up and be given more responsibility, because people could see that I was articulate, able to communicate effectively, explain problems in business language versus technical. So I think those things really helped to build my (I hate the word reputation but basically, my reputation) within the company and with management.
I just want to say the thing that I love most about working in the cloud and on cloud-related stuff (which is what I've been doing exclusively since 2016 and particularly in the last couple years) has been: I'm so much more engaged with the actual business. For the first twenty years of my career at Dolby, I was basically doing IT. I was in a silo working on IT stuff, I had very little interaction with what the business was doing, or only peripherally like, "Ok, let's go build an office" Ok yes, I'm helping the business that way, but for the last two years I've been very much embedded in the business and helping them with their cloud projects because Dolby, as a company, is trying to get into kind of cloudy initiatives. So being able to help with business development, as they go to build these cloud businesses, has been really exciting and fulfilling compared to what I've done in the past. I really feel like I'm actually contributing to the whole company, and not just to IT.
So something I would say is: if you can find a way to add value to your business and not just be somebody in IT off working in a corner that's a big way that you're going to move your career forward is understanding. Understand what your business does and how they make money, and how they can succeed. If you don't know those things and can't talk articulately about those things you're probably not going to be as successful as you could be.
I think from a future-forward looking thing, I don't know that I'm great at that, I think the trends are out there around automation and this notion of infrastructure-as-code is something's been talked about a lot for the last couple of years. I still have a gut feeling that many, many, many companies are still doing ClickOps where they're just going into the console and they're clicking things, and everyone has great aspirations to have infrastructure-as-code, my company included! I would definitely say that we're still on the learning on the beginning slope of that, as far as becoming more expert at that stuff.
I think that this notion of FinOps is particularly interesting. Something that I never even heard of it a year ago and then three months ago I was at the FinOps conference, and meeting with people. Where it's a really interesting mix of engineers and cloud people, and then straight up finance people, and how we're coming together and trying to build this discipline around cost optimization and cost control of cloud, and kind of understanding Cloud Spend, which it can be quite challenging. So I think that's that's an area that I think is going to grow quite a bit, especially as the economy gets a little tighter. I've noticed in the last six months within my own company that there's a lot more focus being paid to cost optimization of cloud spend versus a year ago - I couldn't get anybody to talk to me about spot instances or reserve instances or savings plans; no one will want to talk about that because they're too busy building. Now they're coming back to me saying, "Okay so what was that thing you were telling me about savings plans? And how can I save money on my compute?" So it's definitely a trend that I think is going to continue. I feel like sustainability is going to be a great lever for FinOps in the sense that people may not care about the cost, but they're going to care about the fact that they're using more energy than they need to.
I'm not a big multicloud guy. This whole notion of multicloud, or I think this new thing I've started to see now in the last couple weeks is supercloud. Have you heard this term supercloud? It's this notion of stitching together your clouds into this one landscape. I don't know a lot of people who are really doing that in a real sense. In my company we use a lot of AWS and then maybe we have some things that we're doing in Azure and some things we're doing in GCP, but it's very application-specific or project-specific. It's not like we're trying to build this massive run-in-any-cloud Landscape. Now, maybe there are companies that are doing that. I'm sure that there are, but I feel like that is one of the most overhyped things and a lot of it is companies trying to sell you products to manage your multicloud. If you look at like VMware. VMware is a very interesting company, I love them. I was there from the beginning doing virtualization with them, but they've had to adapt, and the way that they're adapting is, "Ok, we're going to be a multicloud company and help you manage your hybrid cloud or your multicloud. And that's interesting play. I can see why you're doing that, but I don't know whether I really care.
IT in my opinion, is always learning, always growing. If you're not, you're going to be yesterday's news essentially.
And so for me, it's always about what's on the horizon. What can I learn that's new? I knew very little about Cloud Fin Ops six months ago, but for the past six months I've been heavily doing a lot of work around for Cloud financials and really getting more and more into that because the company's required it of me. And so that's just it; you have to go where the where the winds are taking you as far as your career, and be willing to shift gears and do new things. I find that if you're willing to do that and management recognize that you're gonna, that's going to serve you well versus if you just want to be, "I want to be the exchange guy. I want to be the exchange guy for the rest of my career." Okay, you can do that, but you know, you may find yourself at some point out of a job.
At one point I was running the entire infrastructure team. So, all the server guys, all the network guys were under me, and at some point I decided that wanted to get back to being an individual contributor. I enjoy management, I think I'm relatively good at being a manager, but what I told my boss and what my line was like, "That's not where I bring the most value to the company. Where I bring value is in this kind of architecture, researching new things, understanding systems, and bringing new things into the environment. That's my strong suit, not necessarily doing performance reviews and all that kind of junk." So again, my manager at the time supported that and was able to move me back into this individual contributor role that I'm in now and allows me to do what I love.