text stringlengths 0 2.29k |
|---|
**Gerhard Lazu:** I find that really interesting, because it is a world where I used to be a long time ago, and I always thought that this is what improvement looks like. But what I didn't realize is that a lot of people saw things differently. And people that even I know. And that's why I thought this conversation wou... |
**Lars Wikman:** It's a tricky one. I think I'm still a bit in exploration there... Because what my current day to day production looks like is pretty dominated by what my current client is. So I do consulting, and run a team for a product that's being developed at one of my clients. And they are we run things on Fly. ... |
\[06:19\] And I will say, I've been pretty happy with Fly. It has been a mixed bag, because this -- it's still an early company, it's still an early platform. So definitely sort of a mixed experience. But they essentially do what a Kubernetes type solution would do for me. They do platform engineering, so I don't have ... |
So I rolled with whatever was there, and made some choices based off of that, based on what I see... Sort of "Oh, the team's experience level is about yay-high. Okay, we should not spend a ton of our time on the server. Someone else should deal with most of the ops." |
If I needed to get something off the ground on a budget, or if I built my own SaaS, I think I would probably set up a dedicated server for it, potentially to have the failover. It depends a little bit on the service. Not everything needs to be highly available, really... And in that case -- right now I'd probably pick ... |
And there's a big difference between doing Elixir and when I was doing Python. Because if you were doing Python, and you set up an app server, you absolutely should put NGINX in front of it, because that app server was never intended to meet the world. But when you're dealing with the Erlang VM, and well-established se... |
**Gerhard Lazu:** \[09:45\] I really like this -- it's not minimalistic, it's almost like it's a very simple approach to running something in production, that you know can handle the load really well. And I think that's really important. I think Go can be a little bit like that, for people that run Go. it can just term... |
Other programming languages are a bit more complicated. You mentioned Python. I know Ruby, from personal experience, is one of those two. |
**Lars Wikman:** Node... |
**Gerhard Lazu:** Node. |
**Lars Wikman:** Node is super-fast, but it is fast in a very particular way. |
**Gerhard Lazu:** Oh, yes. Oh, yes. |
**Lars Wikman:** And scaling Node can be challenging, and there are footguns with regards to CPU-bound loads in Node, that Go and Elixir have designs to prevent. Or at least as far as I know, Go has that. |
**Gerhard Lazu:** Oh, yes. |
**Lars Wikman:** I know Elixir does. Or Erlang. |
**Gerhard Lazu:** And there's like another up and coming one, Rust, which is even more efficient from a memory perspective, from a CPU perspective. So there are certain languages that are encouraging a simpler operational model. And I think that is something important that many people miss, because they are wondering, ... |
So trying to close this loop, I'm wondering, how much do you think is your choice of Elixir in production down to the simplicity that you can pick? |
**Lars Wikman:** I certainly think my choice of Elixir influences how I want to set things up a lot. Whenever I explore sort of doing multi-language ecosystems, or it's like "Oh, yeah, I need this tool from this language, and this tool from this language, because it would be infeasible to reimplement", then the ops sta... |
There's also -- depending on the work your machine is doing, Elixir has a machine learning project now. And if you're doing machine learning, of course, that really affects your ops, and it could be a good reason to use this sort of Elastic Cloud Service, because if you want to train models on really fast hardware, you... |
**Gerhard Lazu:** Yup. |
**Lars Wikman:** But it is the de facto language for machine learning and AI, and that has nothing to do with Python as a sort of implementation, because everything, every bit of it ends up going to C or C++ to perform. |
**Gerhard Lazu:** Exactly. Yeah. |
**Lars Wikman:** Or Rust, I guess... I don't know if people write bindings in Rust under Python these days. |
**Gerhard Lazu:** Have you looked into Rustler? |
**Lars Wikman:** I'm familiar with it. Since I don't write Rust, I have no reason to poke it further. But yeah, Rustler and Sigler allow you to do Rust and Zig in Elixir. Good stuff. |
**Gerhard Lazu:** So what does your deployment artifact look like? When you push code into production, what are you actually getting out there? |
**Lars Wikman:** \[14:04\] So my ideal is the Erlang release... So as close as you can get to a Go binary, because I think Go does a better job of that than Elixir and Erlang. But there are reasons why Elixir and Erlang do it the way they do. And that has to do with a lot of interesting sort of operational capabilities... |
But yeah, something pretty static that you can just ship over to the machine - I'm not a super-fan of containers. They're super-important if you want to treat everything the same, and if you want to sort of package for a larger ecosystem. It makes good sense to use them then. But shipping a Go binary inside of a contai... |
**Gerhard Lazu:** Okay. So if you're not using containers to get those Erlang releases out there, what do you use? |
**Lars Wikman:** Oh, you know, FileZilla, copy-paste... No. |
**Gerhard Lazu:** \[laughs\] Rsync? Come on... You have to use Rsync. |
**Lars Wikman:** No.. WinSCP, you know...? |
**Gerhard Lazu:** Oh, yes. I remember. |
**Lars Wikman:** No, it would vary... And right now, I don't ship a ton of things to my own servers, but generally, it's just SCP when I do. The things I run for myself, for low-scale production, that's just SCP over And as I mentioned, for clients, I'm currently doing Fly, some Fly deploy command, which does ship con... |
**Gerhard Lazu:** Yep. I was thinking about that. So do you build a container with Fly, or do you let the Fly CLI just figure it all out? Use build packs, and... |
**Lars Wikman:** Our CI/CD builds the container. |
**Gerhard Lazu:** Okay. |
**Lars Wikman:** So we have -- that's something that I grabbed from our conversation about k3s and your demo of ArgoCD. I really like trying to pin down sort of "This is what we're running on this environment, and this is what we're running on this environment." I haven't quite gotten it to the point where it's all def... |
**Gerhard Lazu:** Yeah, that's right. Yeah. To be honest, containers are, in my mind, great as a distribution mechanism. It is a standard distribution mechanism. If you had to ever deal with Deb packages, or RPMs, or tar.gz's - whatever, it doesn't really matter - it is a standard distribution mechanism for code, and y... |
**Lars Wikman:** Swear a lot... |
**Gerhard Lazu:** Okay, we have to start with that, right? "Dammit!" Table flips... |
**Lars Wikman:** \[18:01\] Yeah. So generally, the most important parts are to make sure to have a backup strategy in place, and some kind of disaster recovery. And one thing I've tended to do with my backups that I set up on my ad hoc servers, it's like "Oh, I need to set this up, and set this thing up" etc. It's like... |
**Gerhard Lazu:** Very important, right? |
**Lars Wikman:** So that's sort of the simplest disaster recovery approach. And it's similar to what I would do for a production project where I'm running dedicated infrastructure, and sort of having serious customers, and all that. But there would be more things for more serious projects. For example, I need to find o... |
But I haven't been thrilled with Grafana. I think my best sort of APM-ish experience was when I worked on a product that used New Relic, and this was a number of years ago... But just because it really did give good insights, and then they charged out the nose for it. I think Datadog is probably on the list of what I'd... |
**Gerhard Lazu:** Yeah, for sure. |
**Lars Wikman:** "Monitoring went down, and then... Production went down." |
**Gerhard Lazu:** No, no, "Production went down. Let's check monitoring. Oh, dammit. Monitoring is down as well!" |
**Lars Wikman:** Yeah. |
**Gerhard Lazu:** Okay. |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.