DevOps vs SRE vs Platform Engineering | Clear Big Misconceptions
HTML-код
- Опубликовано: 28 сен 2024
- Get a Free System Design PDF with 158 pages by subscribing to our weekly newsletter.: blog.bytebyteg...
Animation tools: Adobe Illustrator and After Effects.
Checkout our bestselling System Design Interview books:
Volume 1: amzn.to/3Ou7gkd
Volume 2: amzn.to/3HqGozy
The digital version of System Design Interview books: bit.ly/3mlDSk9
ABOUT US:
Covering topics and trends in large-scale system design, from the authors of the best-selling System Design Interview series.
The primary goal of DevOps/SRE/PlatformEng is to make Yaml the single programming language everyone must use for everything.
Good luck with those indentations.
The only thing worse than Yaml is XML.
Use CUE
Ha! Ew.
I am a YAML engineer! :(
@sahn: I think Platform Engineering could have explained in a better way. Below is my simple thought process
Let me clarify Platform Engineering in a simpler way. Imagine it as building the foundational infrastructure for various parts of the organization. Netflix is a good example of this. The main difference between Platform Engineering (PE), DevOps, and Site Reliability Engineering (SRE) is in their primary responsibilities.
PE's primary responsibility is creating the infrastructure that multiple business units can use.
DevOps focuses on delivering applications to production quickly and efficiently, making it easier for developers to deploy their work.
SRE's main responsibility is monitoring and scaling applications to ensure great service for customers and achieve high SLA (Service Level Agreement) targets, like 99.99% uptime.
Engineers in these teams may switch between roles and wear multiple hats. For example, a developer could write unit test cases or contribute to other aspects of the engineering process
Thanks! Well put imo!
At the end of the video, I was thinking what you were thinking; but I knew someone had beaten me to comment! Thanks!
I did not understand this video, however your explanation helped me get clarity. Thank you very much @suru1432002
My company is abandoning Devops, because what ended up happening was the dev's still didn't want to do Operations, they just wanted to quickly deploy code without anybody getting in the way. The result was the devops driven solutions were the most buggy and unsupported components in the stack. Nobody monitored or supported the solutions, leading to frequent production outages. Were going back to separate dev and operations teams with clear responsibilities.
Brutal, hope everything works out.
I was a Linux Admin a few years ago, now working as a system engineer managing tools.
The mindset of a system administrator is very different, You need to do boring repeated tasks like sending mails, audition infrastructure and monitoring it for maintaining good infrastructure.
Developers, Devops mindset don't work like that, they want to build code fast, infrastructure fast, automate it and fix it fast if they have a problem. They are least worried about the whole system and how to maintain it.
Administrators regularly checks infrastructure, send mail to users to know why users are using services, maintain regard to who is getting access, mantaing records on who is making infrastructure changes, Monitoring, pulling reports to know inventory, access, unwanted resources, need for changes in infrastructure, building process to make jobs of everyone easy
Thanks for sharing your experience. Could you reveal a bit more without giving away any proprietary information? Like why exactly was no one monitoring?. What is the type and scale of your software? Simple crud operations? Streaming? Gaming?
"SRE = Civil Engineering for software" ---> great analogy
4:07 Great punchline, which bears repeating! Thanks, bbg! 😎✌️
Too many terms. To simply, DevOps is a modern org mindset of delivering software and services to customer.
SRE is a set of practices dedicated to Production reliability. Platform Engineering is the art of serving consumers with the platform in scope. Platform can be Tooling, services, apps or just infra. Consumers can be internal or external.
DevOps idea is to break silos between Ops and Dev engineers. Breaking silos means involvement of both camps from planning to delivery and maintenance.
A squad who is responsible for developing, running and maintaining Service A means it is composed of engineers with different backgrounds (dev/ops) working together for Service A. The Ops engineer in this squad are modern Ops people who came from developer background and transitioned to Ops, therefore can automate, code, integrate and operate. These modern Ops must understand the context of most pillars of software engineering. Example:
-they can get productive in a week to use a PL they never used before.
-they can get productive in provisioning cloud resources in any cloud platform
-they are generalist, they don't specialized in specific solutions, instead they select solutions specific to situation.
-that is why these people are highly EXPERIENCED and understands that environments are all the same until the purposes are defined.
Do you guys have something like discord or slack for question and answer?
Brown Jason Davis Matthew Brown Sarah
I wasn’t even confused before watching this video but now I am.
Consider a situation where the world has widgets and they want to convert them from blue to red.
SWE: Let's build a micro-service that converts widgets from blue to red and a test plan to validate the process.
Platform Engineer: Let's build tools/databases to manage all orchestration and storage needs.
DevOps: Let's build the high level cloud services like permissions.
SRE: Let's make sure updates from SWE's pass the test plans then deploy those updates.
There is a fundamental disconnect in two of the main definitions of DevOps. "You build it, you run it = DevOps" *removes* operations entirely, while "Dev + Ops = DevOps" integrates their workflows by removing the barriers that silo'd them in the first place.
Cluelees people at big tech once again bastardizing terms. Software engineer another example
No, if you look back at his diagram, the "You build it, you run it" philosophy was based on teams. A team will and should consist of operations members in DevOps
@@phoenixamaranth No, they literally mean that the individual software developer that writes the code also has to run the code in production and takes turns being on-call to fix problems when they arise at 3 am.
@@DemPilafian No, they literally don't. Again, you willfully ignore the team and that the team consists of operations members in devops. Do you even work in software development? You'll end up on call no matter who deploys the software you built. That's part of the job sometimes.
@@phoenixamaranth You're talking about faux-DevOps. It's very common -- just like it's very common to have "Agile" shops that aren't actually Agile.
This is one of the best explanations I have come across recently regarding DevOps, SRE and Platform Engineering 👍👍
I work on a backend system. DEVOPS is a con. Whatever it was meant to be in reality companies only focus on the alleged speedy development. I have yet to see it deal with the reality of a production service in an industry which has to adhere to layers of compliance such as security and audit.
While it does, or should, help in unit testing what I have found are projects which are becoming more skilled with little understanding of how their work interfaces with other systems.
The less said about Agile the better. Agile is like building a plane and trying to take off at the same time. All while defective parts are being replaced. What could go wrong?
C# programming lessons to come👍♥
Most company do Scrum + DevOps, which means there is a DevOps engineer as part of the team. This person is basically a system administrator with cloud knowledge, usually AWS, less frequently Azure and GCP. He can't write a program and he has no idea when to use DocumentDB and when DynamoDB. The issue with this approach is that only one person knows how to deploy and running the services. To deal with this issue, sometimes companies create a DevOps team, but this often does not work either, because it becomes a bottleneck.
Sometimes developers need very specialised tools, like Kafka, Flink, that DevOps simply can't scale or operate properly, due to the massive amount of configuration required. So for me it's way better if there are no DevOps at all.
Facebook instead employs SE in teams called Production Engineering Teams. Basically these are skilled developers with deep knowledge of the operating system and application they run. People who can debug the code. Facebook does not use Kubernetes or ECS or whatever, so these teams ensure a service, or a bunch of services, are up and running. But do not make mistakes, these are not DevOps, these are skilled Software Engineers.
What I personally did, as tech lead and CTO, was to create a platform. In small startups I created the platform myself, and then developer used it. My colleagues were surprised we were able to cut 80% of the code. In bigger companies I usually create a team, that I call R&D, but it's basically PE, that is responsible to build the platform.
I do not think one solution fits all use cases. But PE for me is better, because the entire team acquires knowledge of the platform using it. They learn how to deploy, how to make improvements to the platform itself. I am a huge fan of PE and monorepo.
Interesting. What skills in your opinion would a DevOps engineer need to learn to become a PE ?
@@IkechiGriffith oh, the gap is ginormous. You need to learn a lot about system design and patterns, and gain a significative experience designing systems. Also, I don't think you can be a PE if all you have done is writing bash or Python scripts. You need to be able to write complex software and master at least a programming language.
This is so spot on. Lately, companies are looking instead for "Software Engineer, Site Reliability Engineering." They want to make sure the SRE they hire have very strong software engineering skills.
I am a bit curious about the 80% cut in code. Forgive me for being sceptical. But I would love it if you have a blog/video where you explain how to do this. I understand that I may not be able to replicate your success rate but I would love to learn from your example .
ByteByteGo - please please share the tools and softwares you use to create these wonderful videos. It will be extremely helpful to learn them and use it for work and share knowledge in general. Thanks in advance!
I have no words I'm confused with these concepts from the past few days thanks for the knowledge that you share💌
Nonsense
What is difference between platform engineering and cloud computing please explain both are same or one different names please my doubts please reply my comments
Thanks always. They really do all have the same Objective, sustainably delivering value.
basically business need to speed up delivery and guarantee service
are thee any print worthy copies of the ~slides available online ?
is this a new video or republished?
Great explanation
Keep going!
Platform engineering still not explain to understand easily .
when this channel is known to explain concept in simple way 😮.
Truth is, DevOps, SRE and PE is 99% the same but people invent new titles to feel more special
@@PouriyaJamshidiand for higher salary 😅
@@y0d4 and that 😄
Now I am understand....
It's Salary and Title increment Game.
@@PouriyaJamshidi Nailed it. Whatever these terms are "supposed" to mean, the truth is that an SRE at company A could be performing the role of a DevOps engineer at company B.
worst explanation
which software you use for making this type of videos? @bytebytego