Big Ruby 2014

La infraestructura tecnológica de GitHub

Ben Bleything  · 




Extracto de la transcripción automática del vídeo realizada por YouTube.

sorry about that my name is ben and i work at github and i work in the ops group at github and my timer didn't start so sorry about that i work in the ops group and specifically i work on supporting and maintaining and building infrastructure for our internal

applications you may have heard other hubbers talking about things like team which is our internal kind of twitter type social networking site or we have a thing called higher where we track candidates and interviews and we have a whole bunch of those we have

about a hundred and fifty applications that we use internally they're not all web apps some of them are utilities and services that enable other things but there's a lot of them and my my team supports all of them most of the internal tools infrastructure

is on ec2 in fact all of it is on ec2 either directly VMs that we own or via Heroku and that's what I I want to talk to you about today I personally just as a technologist I really love the cloud I think it's a super cool concept I actually know what

it means I don't use it as a marketing term and I think it's awesome I really like the idea that i can just say hey i need twenty-five cores right now for 10 minutes and get them and then throw them away and be done with it I love that I love the elasticity

and the on-demand nature of compute resources and it's just super cool it's really exciting to me our use of AWS and Heroku at github isn't really anything special almost everything that we do is stuff that we've heard other people doing and

copied we take some notes from netflix they're huge AWS user a lot of our ops team came from Heroku and so they know the right ways to interact with Heroku so none of it i don't i don't think any of it is groundbreaking but i still think that there's

a lot of stuff in there that's kind of cool and a little bit different and and that's kind of what i wanted to share with you today my hope is that by the end of this you'll be able to come up to me and tell me what I'm doing wrong what things

we can do better and and where we can improve and then maybe also at the end of it go and build some of the things that we haven't gotten to yet and then open source i'm so i don't have to build them i'm going to be talking about chat ops a

little bit if you haven't heard this term before it's how we refer to our way of working github is a little bit more than sixty percent distributed meaning not in San Francisco in fact nobody from the ops group is in san francisco at all if you look

at the ops group I think there's 16 or 17 of us and the largest concentration of ops people in a single City is three there's no city that has more than three office people in it and so whereas github is very focused on using chat and campfires what

we use as our primary method of communication the ops team tries to use it as our primary method of work as well and that means automating all of our common tasks common operational tasks through Hugh bot and I'll show you some examples later but if you

want to get a really good background on this go look up Jesse Newlands talk from Ruby fusa last year 2013 it's on YouTube it's a great talk and a really good introduction to the concept of chat ops in the context of our ops team for a sense of scale

I'd like to kind of give you an idea of what our infrastructure looks like the numbers here are deliberately vague and I apologize for that but I've been told that I can't tell you real numbers so we have a data center it is in the middle of nowhere

in Virginia which has some advantages that i'll get to in a second we have some number of hundreds of machines there it's more than a few hundred and less than a thousand they're mostly Dells they're mostly CSeries high-density sled servers

if you're a hardware nerd like I am come find me later I'll tell you all about them because i think the CSeries chassis is are super cool the data center powers pretty much all of github com when you go to the website most of what you're seeing

is being served out of the data center there's a few systems here and there that are on ec2 but but not many interestingly we do almost no virtualization in the data center the what we do is in service of our continuous integration systems and some of

our build systems and no virtualization is at any point involved in the serving of calm which kind of surprises me every time I think about it but that's what we do one of the advantages of being in the middle of nowhere in Virginia is that we are figuratively

right next door to the u.s. East one region of AWS which is kind of cool because it means our latency is a function of the speed of light I mean I guess it always is but like the we're very we're very very close to us East one and I'll talk about

in a few minutes why that is as cool as well cool enough to justify how excited I am about it anyway we have instances and resources in five regions of AWS they're primarily newest east one but we also have stuff in Europe in u.s. west and a couple in

[ ... ]

Nota: se han omitido las otras 2.460 palabras de la transcripción completa para cumplir con las normas de «uso razonable» de YouTube.