My best tweets and threads about "Platform Engineering" from 2022 ♻️

With bonus thoughts and commentary 😁

🧵 👇 This was my original thread on Platform Engineering back in February that really kicked off (I think this is my most popular tweet, ever!) 🚀

https://twitter.com/danielbryantuk/status/1494614250567966732

I've had several great chats with folks looking to get into devrel this week, and wanted to share my current advice for the role of developer relations, developer advocacy, etc 👩‍💻 👨‍💻

Spoiler alert: it's not just developer marketing 💸

Read on 🧵 👇 My general take on devrel is that you're bridging the gap between technology and user adoption 💻 🌉 🧑‍🤝‍🧑

- Generating awareness 📣
- Education 🧑‍🏫
- Creating tech demos 🛠️
- Community building 🧑‍🤝‍🧑
- Empathizing with devs 🧰
- Acquiring feedback 👂

What is HTTP/3 and why should you care? 🤔

- Reduced latency and buffering
- 70%+ browser support
- Benefits for machines with old OS/kernels

My team and I have been chatting to experts and playing with the latest version of the protocol that powers the web

Read on 🧵 👇 HTTP/3 is supported by 70%+ of browsers (including Chrome, Firefox, and Edge), and according to W3Techs 25% of the top 10 million websites 🌐

The official RFC/spec was standardized in June 2022:

datatracker.ietf.org/doc/rfc9114/

During my travels I listened to three very good audiobooks for folks working in the dev tooling/platform space:

- Inspired, by @cagan
amzn.to/3bmpiqt
- Ask Your Developer, by @jeffiel
amzn.to/3A5BqXm
- The Power Law, by @scmallaby
amzn.to/3Omtq8y

🧵👇 I've read Inspired once before, and so this was a great reminder of what it takes to design, build, and lead teams to make great tech products 🧑‍🤝‍🧑

For folks building cloud platforms, this is a super useful reference for making sure your thinking is product-focused

The latest @InfoQ DevOps and Cloud InfoQ Trends Report has been published 📊

Key topics:
- Data Observability
- FinOps
- eBPF and WASM
- Supply chain security
- Low/no-code platforms
- "Developer Experience as Decision Driver”

Read on to learn more 🧵 👇 @InfoQ Every year, several @InfoQ editors and members of the community get together to share their opinions on the current state of the DevOps and cloud space.

We aim to track the diffusion of technologies, methodologies, and topics in this space

Here is our current take:

5 technologies that #PlatformEngineers should pay attention to for the next six months 🧰 📆

- devcontainers
- Buildpacks (and SBOMs)
- "Remocal" dev tools e.g. Skaffold, Telepresence
- HTTP/3
- Service catalogs e.g. Backstage

Read on to learn more! 🧵 👇 devcontainers enable the packaging and encapsulation of developer environments and can be run locally and remotely 🛠️

This @InfoQ article from @avdi (reviewed by @BeardedCoder) is super useful for establishing the benefits of devcontainers

infoq.com/articles/devco…

Key learnings from the inaugural #PlatformCon

- Everyone is building a platform, whether they realize it or not
- Platform as product FTW 🚀
- Getting exec/C-level buy-in is challenging
- Platform Product Manager will be a hot role 📈
- Balance devex, CLI, UIs, & IaC ⚖️

🧵👇 1/ Everyone is building a platform, whether they realize it or not 🤔

There was a real mix of platform tech at the event:
- Kubernetes
- Serverless
- On-prem

One key theme: every "platform" helped devs to code, ship, and run, reducing toil and increasing speed and safety 🚀 ⛑️

#KubeCon EU Takeaway 2: ➡️ Platforms and “golden paths” enable productivity and reduce developer friction

I saw several great talks, and also presented on why I think golden paths and platform engineering are the next big thing 🔥

Let's explore this in more detail: 🧵 👇 The first mention of "golden paths" was from @MercedesBenz in the opening keynote, where they talked about the K8s-based platform they had built over the past 7 years 🏎️

Lots of thought, effort, and iteration had been applied to the dev experience 🧰

https://twitter.com/danielbryantuk/status/1526836683798368256

My top five takeaways from #KubeCon:

1. Cloud education is vitally important
2. Platforms and “golden paths” enable productivity
3. Developer experience is top of mind
4. Cloud networking is simplifying
5. Increasing focus on security

Let's explore 1 in more detail 🧵 👇 First, the complete #KubeConEU summary blog post that provides in-depth coverage can be found here:

blog.getambassador.io/kubecon-eu-202…

My top Twitter threads on cloud, containers, and microservices for this year so far:

- Platform engineering
- API gateway vs service mesh
- @Kubernetesio debugging
- @Docker chaos testing
- Fast dev/test with Telepresence
- @buildpacks_io
- Microservice testing

A 🧵 of 🧵 s 👇 I'm predicting big things for the "platform engineering" space over the next year. Whether you ❤️ or 😡 the name, I think this is the new DevOps.

Watch this space for lots of knowledge sharing, innovation, and VC money 💰 !

https://twitter.com/danielbryantuk/status/1494614250567966732

Testing microservice-based systems is hard 😢

With that said, software testing in general is challenging, but to paraphrase the late great B.I.G.: "mo' services, mo' problems" 💰

Here are some key references and thinking points 🧵 👇 I think a lot of testing challenges come with a misunderstanding about coupling and cohesion 🤔

As I wrote back in 2018, these two core architectural properties massively impact your ability to test: danielbryantuk.medium.com/microservice-t…

Are you looking to perform simple chaos testing on your @Docker setup to check your system's and microservices' resilience and security?

For example, do you want to spike CPU, memory, i/o, in a container and see what happens?

My top three tools to get started with 🧵 👇 First, you'll need some basic visibility/observability into what's going on.

I've recently been using @bcicen_'s excellent "ctop" (shown in my first tweet): github.com/bcicen/ctop 📊

Once I've seen something obvious, then I run "docker container top 20fa446aebf..." 🔍

Looking to get fast feedback when testing microservices, but can't run all the services on your local machine? 🤯

Are you using @kubernetesio and stuck in a slow build-push-test cycle? 🐢

Let me show you the power of the new (beta) @Docker Desktop Telepresence extension 🧵 👇 At last week's @Docker community all-hands, @scottcjohnston announced the new beta Docker Desktop extension functionality 🥳

@gtardif followed this up with a great live demo 🙌

You can read more about this in my summary blog post: blog.getambassador.io/happy-birthday…

Currently thinking about how API gateways and service mesh can work together, now and in the future:

- North-south traffic
- API as a product
- Migrating to the cloud
- Zonal architecture vs zero trust

A thread 🧵 👇 Traditionally API gateways have handled north-south (ingress) traffic and service meshes have handled east-west (service-to-service) communication

We used to think external access and choose an API gateway. And recently we used to think internal access and choose a service mesh

I've recently been seeing increased chatter about "internal developer relations" or "internal devrel" roles 🤔

I'm also bumping into this role in companies we work with, alongside "dev enablement" and "developer experience (devex)" teams

Want to know more? Read on 🧵 👇 Having worked in externally-facing devrel roles for a while, this is my definition of devrel

DevRels bridge the gap between technology and user adoption:
- Generating awareness
- Education
- Creating tech demos
- Community building
- Empathizing with devs
- Acquiring feedback

In preparation for fixing broken Kubernetes clusters live on @rawkode's #Klustered event, I reminded myself of some of the core K8s debugging commands and techniques

Here are my top 10 tips for platform engineers debugging Kubernetes and the machinery underneath the covers 🧵 👇 First off, use kubectl to take a look at the cluster infra:

$ kubectl get nodes
$ kubectl cluster-info dump

These commands typically give you a good idea of where to start debugging, e.g. broken nodes, infra issues, resources

"Platform Engineering" is rapidly becoming the new DevOps or SRE. Almost every day we hear about another org building an internal developer platform or control plane.

Want to know what platform engineering is, where the trends are going, and why you should care?

Read on 🧵👇 We've all been building application/web platforms for years

- On-premises: ticket-driven, bare-metal, long lead time
- First-gen PaaS: self-service, VM-based, one-size-fits-all, on-demand
- Next-gen PaaS/Custom platform: self-service, container-based, fast feedback, good UX/DevX

My top 8 commands and tools for debugging applications running on @kubernetesio 🧵👇 A good first step is viewing the app's pods and associated logs (potentially targeting a specific container) looking for obvious crashes

$ kubectl get pods -n my-namespace

$ kubectl logs my-pod -c my-container

More info -> kubernetes.io/docs/reference…