#187 — October 10, 2018 |
Web Operations Weekly |
Facebook StateService: Automating Recovery of Third-Party Services After Outages — StateService, now open sourced, is a system Facebook uses to restart VM-hosted third party services after outages without having to rely upon manual effort. GitHub repo. Facebook Code |
Google GKE vs Microsoft AKS vs Amazon EKS — A comparison between the Kubernetes deployment and management services offered by three major cloud platforms. Steven Acreman |
Deliver Web Assets Up to 70% Faster with Spaces Object Storage — Spaces object storage now includes a built-in CDN, providing global edge caching to improve user experience. DigitalOcean sponsor |
Netlify Raises $30M For Its 'Application Delivery Network' — The company have engineered a new platform “where content and applications are created directly on a global network, bypassing the need to ever setup or manage servers”. Netlify |
Buildpacks Go Cloud Native — Heroku and the CNCF are collaborating on the development of ‘Cloud Native Buildpacks’ (CNB) that turn source code into Docker images so developers don’t have to focus on setting up the environment in which their code runs. Heroku |
661 TBit/s, Efficiently, Over A Single Fiber Optic Cable — The achievement, which is solely in the domain of research for now, sent more traffic than currently used by the entire Internet over a single cable. The key achievement, though, is the technique used is particularly energy efficient. Alex Ronin |
Stack Exchange is Migrating to .NET Core — While it’s sometimes claimed the .NET Core platform isn’t mature enough to port over large, existing code bases, the creators of Stack Overflow are ready to go. Microsoft have also written on the future of .NET here, including explaining how the .NET Framework and .NET Core relate to one another. Nick Craver (Stack Exchange) |
💻 Jobs |
DevOps / Platform Engineer - Kubernetes, Cloud Foundry, BOSH — Accelerate your experience & build PaaS products that help developers to ship software faster. Anynines |
Join Our Career Marketplace & Get Matched With A Job You Love — Through Hired, software engineers have transparency into salary offers, competing opportunities, and job details. Hired |
💬 Stories & Opinions |
Managing Data Store Locality at Facebook Scale with Akkio — A look at an interesting system Facebook has built to efficiently store data at a granular scale (in ‘microshards’) across the globe without having entire replicas in each datacenter. Facebook Code |
DNS Security Remains An Issue in Tech Industry — Here’s a lay of the land along with some proposed solutions. Larry Seltzer |
@here, Is Anyone Using Staging-1? — Check in and out shared resources in Slack and stop interrupting co-workers when you need something. checkt sponsor |
Validating Performance and Reliability of the New Dropbox Search Engine Dropbox Tech Blog |
How We Store Data in the Cloud at Auth0 — MongoDB, PostgreSQL, and Redis all have a role to play at the popular authentication-as-a-service provider. Dirceu Tiegs |
Risk Detection Infrastructure at Postmates — Postmates is a logistics company which has payment-related risks to manage. Ayush Bharadwaj (Postmates) |
📘 Tutorials |
How to Better Watch Your AWS Costs Before You Forget — No-one wants to get a nasty surprise in their AWS bill at the end of the month. Here’s one approach to tracking things. Kyle Galbraith |
Service Imports Make Config Management Easier Than Ever Manifold sponsor |
Using Serverless on GCP to Add Custom Intelligence to Gmail — A look at what’s involved with building Gmail extensions with Cloud Functions and other fully-managed services on GCP. Google Cloud Blog |
Patterns for Distributed Transactions Within a Microservices Architecture — Managing distributed transactions across multiple microservices poses numerous challenges. Keyang Xiang |
'What I Learned From The CI of Facebook and Their Open Source Projects' — 4 tips on how to better use a continuous integration system in order to focus on what matters instead of busy work. Aurélien Le Masson |
How to Manage Prometheus High-Availability with Postgres and TimescaleDB — Prometheus is a popular open source monitoring system. Niksa Jakovljevic |
How to Secure Your Server Room Steven Vaughan-Nichols |
🔧 Tools |
PostGraphile v4: A GraphQL API for Your PostgreSQL Database — Execute one command (or mount one Node.js middleware) and get an high-performance GraphQL API for your database. Benjie Gillam |
RoadRunner: A High Performance PHP Application Server — Handles load balancing and process management and can be used to replace a classic Nginx+FPM setup. Spiral Scout |
HTTPS-PORTAL: An Automated HTTPS Server — Powered by Nginx, Let’s Encrypt and Docker. Weiyan Shao |