text
stringlengths
0
2.67k
**Pete Naylor:** Yeah. Yeah. That was that was all Amazon, actually. Well, I worked at these companies and it was mostly about systems admin stuff. I've never been a developer. A lot of my stuff that I used to do was building packages for automated installation on all of these servers. They were mostly C. There was als...
**Justin Garrison:** And this was the '90s, or 2000s? Were you doing like spec files? What was this?
**Pete Naylor:** '90s and early 2000s. So that company --
**Justin Garrison:** Okay. It was just TARS.
**Pete Naylor:** Yeah. A lot of TARS. Yeah. Unpackaging things, running the builds, make files... Yeah.
**Break**: \[21:47\]
**Autumn Nash:** How'd you get into like Dynamo, and like all the different NoSQL, and all that? Because Pete's really, like really, really good with all of that. But Pete has got that -- Pete won't bullsh\*\* you. You know what I mean? A lot of people will be like "I sell this thing, and I'm just going to say it is th...
**Pete Naylor:** I still have that little bit of like where I grew up in Australia.
**Autumn Nash:** That's why I love you, though.
**Pete Naylor:** My hometown, there is no cooth. Nobody holds anything back.
**Autumn Nash:** Look, we were in the room with some of the most popular-ish database nerds of like all of the tech people, and a lot of people are scared to like disagree with people sometimes, and like they'll be like "Well, this person is so good at this, and they invented all these things." Pete will be like "Look ...
**Pete Naylor:** So I developed through that. My company got bought out, and all of that stuff happened. And eventually, it was like Verio. I don't know if anybody remembers them, but they were this company that bought up 50 of the most successful ISPs all around the US. And when they did that, they were like "Well, we...
**Autumn Nash:** I think that's just like the cycle.
**Pete Naylor:** So I did that, and eventually I kind of got into these roles where I was managing, and looking after teams... And a lot of the folks that I supported were DBAs. When you're running the sysadmin stuff, there's -- usually, when things went wrong, there was a lot of finger-pointing... It's like, "No, it's...
When I joined AWS, I actually applied as a TAM, so a technical account manager. At AWS, that's part of the enterprise support org. And when I joined, it was like in my first week, they took me aside and they were like "Well, we're going to assign you your customer, and your customer is going to be Amazon.com."
**Justin Garrison:** Oh, shoot.
**Pete Naylor:** Yeah. I was like "I don't know what this really means. It sounds kind of cool, it could be fun..." And then I started telling some of the other TAMs about it, and their eyes were like "Oh, man, you're in for a ride. It's going to be crazy", because they're just a huge, huge customer, and very, very dem...
But yeah, I started in that role, and it was a very small team of us that did grow over time... And we had to teach all of those internal teams to like "Please use enterprise support, because this is the only way this can scale."
**Justin Garrison:** And for anyone that's not familiar with how that relationship works - Amazon, the company, amazon.com uses AWS as kind of a customer, where they use the services and they are supposed to go through some sort of support channels for standard things... But it's the largest, most important customer, t...
**Pete Naylor:** Kind of overwhelming to work with them and support them. But it was also really cool. They were really pushing a lot of the AWS services harder than most of the customers. So we got to be involved in things where - you know, there were some sharp edges in DynamoDB, for example. And when we could go and...
\[28:15\] So it was it was cool to be a part of, but very, very stressful, intense. None of the tooling that was there, that all of the other teams used to support their customers, none of them would work for us. They just, they didn't scale. They would fall over. So yeah, I worked on that... So to your point about Ama...
**Justin Garrison:** No.
**Pete Naylor:** No. And at this point, they were still doing things where they would kind of share data centers, and they would rack stuff up... But for example on a prime day, these teams would go on and choose really basic, bare metal stuff and have it racked up, and that's how they were scaling a lot of stuff. But ...
**Autumn Nash:** Oh yeah, that was a big one.
**Pete Naylor:** So I think in my first couple of weeks there had been some major event. One of the teams who had just started using DynamoDB had maybe not figured out how that works, and had some hot key things, and it was a big operational issue, and a review of that. And I got sent off to deal with that, with no kno...
**Autumn Nash:** For context - so going from Oracle, which is a relational, you know, and then going to like one of the biggest, like, no SQL databases... That might not have been the biggest at the time, but just the access pattern, and how you use DynamoDB is very, very, very different than a relational database.
**Pete Naylor:** Yeah. There was a lot to teach. And for the Amazon teams, it's like, you hear about the whole two pizza thing... And that is such a reality that it's painful to try and move them through something like this. It's like, how do you reach out and get all of those people to know about "We're going to do th...
So there was a lot of stuff... It was done by tiers, so anything that was really, really critical, the kind of stuff that Jeff would get paged within an hour of an impact - that was the tier that all had to go to DynamoDB. And then there was a little bit lower tier, where it could stay on a relational database, but it ...
I think the DynamoDB thing really spoke to me most, because of that whole change to the completely serverless model, and the scale. It's super-cool. I got to go in and do things like sit in the war rooms for Prime Day, and help all of those teams trying to plan for scaling... And I think a lot of those teams didn't see...
**Justin Garrison:** \[32:04\] Yeah.
**Pete Naylor:** And so I got to see like how much time that gave back to them. It was pretty cool.
**Justin Garrison:** According to the announcement when Amazon announced that they had moved off of Oracle officially, which was October 2019... So five years --
**Pete Naylor:** I'm in that video.
**Justin Garrison:** You are?
**Pete Naylor:** You can see me drinking champagne. Yeah.
**Justin Garrison:** Yeah, that's awesome. But it says there was 75 petabytes of data, 7500 Oracle databases, and moving them to, like you said, all of the different options, depending on what the right choice was at the time.
**Pete Naylor:** Yeah.
**Autumn Nash:** Which is like why those products are actually like really good managed databases, because they tested them out like with that scale of data. Like, you're never going to get scale like that from just random customers; to be able to find the edge cases of your products with those -- for one, with the req...
**Pete Naylor:** It was very cool to be a part of that. And I think, through that process, as hard as it was, that's where I really -- I drank the Amazon Kool-Aid. I loved the LPs, and I really got to see some of those processes happen from both sides, where I would go to meetings where there were a bunch of people on ...
So I really loved those LPs. I really believed in it because I saw it from both sides of the org. They're slightly different cultures, I think, Amazon retail versus AWS. So I just got really sucked into that. I loved it, and I learned so much. After a lot of that, those programs were kind of wrapping up, and then I had...
So I went and basically worked with all of these other really big companies. We'd each get a few of them, a handful, that was like "You need to guide these folks through this whole modernization journey, and in particular help them to adopt DynamoDB."
**Autumn Nash:** Being a specialist, I'd say, is one of the coolest jobs, because you get to see all the contrast between different databases... And it's the perfect mix of business and technology, and seeing all of that contrast, and watching migrations, and seeing when you need to move from like a relational -- becau...
**Pete Naylor:** I try to -- like, when I first started as an SA, I really liked the idea... Because I never had like a sales kind of background, and I think I'm probably not the right person to have a lot of those discussions... Because I usually just -- I really would just want to solve the problem and make it awesom...
So I did that for a few years. I had a lot of really great experience there, too. And then eventually, the product management team - because I worked so closely with all of the engineers on DynamoDB, and some of the product managers, and I really got to loving some of those folks. Some amazing, amazing engineers within...
\[35:51\] It's big numbers. It's super-cool. I still can't really share any of this kind of thing, but if people could hear the number of nodes and the number of servers and things that are behind DynamoDB... It's just absolutely astounding. It's an incredible service.
**Autumn Nash:** They created real magic in real time. That's technological magic right there.
**Pete Naylor:** Yeah. So yeah, I mean, that was that next stage. And then and the product management team kind of reached out to me and said "Hey, we need more product managers. And you know the product so well, you're really well connected... Have you thought about trying product management?" It's kind of like -- you...
**Justin Garrison:** Yeah.
**Pete Naylor:** And I think at that point I was like "I've seen so many things that customers are asking for, and we've asked for them, and the SAs have asked for them before... It's like, I'm so frustrated that a lot of them are not just getting done. How can this be?" And I thought, "Okay, I can move into this produ...
**Autumn Nash:** What's your favorite job out of all the jobs that you've had?