|
🔗 Stories, Tutorials & Articles |
|
|
|
How we decreased GitLab repo backup times from 48 hours to 41 minutes |
|
|
GitLab pulled a clever move. They swapped out a clunky O(N²) nested loop for some slick mapping, cutting down backup times from 48 hours to a zippy 41 minutes. Your massive repositories now scale better and cost less. In short, they made it faster and easier. Science wins again. |
|
|
|
|
|
|
DevOps: Automating Release Tags |
|
|
GitHub Actions just got a shot of adrenaline. The workflow now slaps tags on releases with spicy semantic versioning. It skims through PR details for those major head-turners and voila—auto-generated changelogs that save time and sanity. |
|
|
|
|
|
|
Systems Correctness Practices at Amazon Web Services ✅ |
|
|
AWS taps TLA+ and P language to hammer out service correctness. Bugs quiver and deadlines whistle past with formal methods wielded like a scalpel. Enter PObserve—this tool is the wizard that conjures log validation magic between design and production. And P? It’s the S3 whisperer, driving sudden consistency gains and cunning optimizations. Formal methods can be a beast, but AWS rides it like a bull, mixing formal proofs with real-world clout. Case in point: a sprightly 94% RSA throughput boost on ARM CPUs. |
|
|
|
|
|
|
You’re not a platform team if you’re just managing infrastructure |
|
|
Platform engineering? It's not just gift-wrapping infrastructure as a service. It's about handing devs the reins and saying, "Go wild." Think of it like an Internal Developer Platform (IDP), similar to the Google Cloud Platform. Here, users truly own their services. The result? Scalability soars, bottlenecks crumble, and devs feel like the rockstars they are. |
|
|
|
|
|
|
Exploiting CI/CD with Style(lint): LOTP Guide |
|
|
CI/CD is vulnerable to Living Off the Pipeline (LOTP) attacks via tools like linters, formatters, build, and test tools—no need to modify workflows. Hacking depends on unexpected code execution, context files, plugins, environment variables. |
|
|
|
|
|
|
Mastering Terraform Variables: Complex Structures & Input Validation |
|
|
Object variables in Terraform crank up your code's modularity. They hand you structured, adaptable input—a game changer for scaling beasts like EKS Nodegroup in the cloud. |
|
|
|
|
|
|
Building Uber’s Multi-Cloud Secrets Management Platform to Enhance Security ✅ |
|
|
Uber built a Secret Management Platform to consolidate vaults and automate secret monitoring and rotation. They use real-time and scheduled scanning to catch leaks early and reduce exposure. They also enable automatic rotation of 20,000 secrets per month on average, prioritizing certain types and platforms first. Additionally, they designed a Secure Secret eXchange system to securely share secrets with third-party vendors without human involvement. |
|
|
|
|
|
|
Shopify Tech Stack ✅ |
|
|
Shopify's stack might look like a minimalist's dream—Ruby on Rails and React. But don’t be fooled; it flexes serious muscle, wrangling 173 billion requests in just one day. They've supercharged Ruby with the mighty duo of YJIT and Sorbet, flung React Native across key apps, and turned to Kafka when sending 66 million messages per second feels like a typical coffee break. |
|
|
|
|
|
|
What I’ve Learned from Designing Landing Zones On Google Cloud |
|
|
Cloud Foundation Fabric and FAST make Google Cloud feel more like a well-oiled machine than a hair-pulling puzzle. They slice through the setup with killer precision, laying down a rock-solid, enterprise-grade foundation. No IAM madness. No network disasters waiting to explode. Just scalable, secure consistency. Your Google Cloud BFFs. |
|
|
|
|
|
|
Are You Over-Engineering Your Tests? – Think Like a Tester |
|
|
Over-engineering alert: Automating every last thing? Recipe for disaster. Flaky tests galore! Stick to manual edge cases and sharp, atomic checks instead of drowning in script spaghetti. Abstraction overload ahead! Chasing too much abstraction makes maintenance a headache. Keep tests clean and clear. Stick with what's working: Chasing every shiny new automation tool only leads to chaos, not clarity. |
|
|
|
|