Bookmarks (18) clear filters

Docker Releases Plugin for Simplified Deployments into AWS ECS and Fargate — Docker has announced the public beta of their enhanced integration of Docker with AWS Elastic Container Service and AWS Fargate.

Red Hat Mandrel Makes Java Native — Mandrel is a new GraalVM-based Java distribution that compiles Java apps directly to native code.

To Microservices and Back Again: Why Segment Went Back to a Monolith — When Segment moved to a microservices architecture, they gained environmental isolation, but at a cost of higher operational overhead. Three years later, the costs were too high, and the team migrated back to a monolith..

Service Mesh Ultimate Guide: Managing Service-to-Service Communications in the Era of Microservices — If the term ‘service mesh’ has ever gone over your head, this is a reasonable primer to what they are, why you might use one, and the terminology around them.

The Status of HTTP/3 — HTTP/3 is the next protocol for network communication across the Web, which is meant to partially replace HTTP/1 and HTTP/2. This a brief but link-packed update on where things stand.

Building Resilient Serverless Systems — A 40 minute talk from QCon London on using serverless technologies and an infrastructure-as-code approach to build, and operate resilient large-scale systems, including a walk through a demo where failure is simulated and recovered from.

Not Sold Yet, GraphQL: A Humble Tale from Skeptic to Enthusiast — Garrett Heinlen talks about how Netflix builds and deploys GraphQL and how they are running it in production.

▶  PID Loops and the Art of Keeping Systems Stable — A developer involved in building EC2, S3, and other AWS services shows what PID loops (essentially feedback and control loops - nothing to do with process IDs!) look like in the context of modern systems, and how exponential backoff, flow-control, and other techniques can be wielded to build self-healing systems.

The Pipeline Driven Organization: Enabling True Continuous Delivery — “The less human bottlenecks stand in the way of pipelines, the more pipeline driven we become, which enables true continuous delivery.”

Athena: Automated Build Health Monitoring at Dropbox Engineering — Dropbox’s engineering team runs 35,000 builds and millions of automated tests everyday so they created a build monitoring system to minimize the manual intervention necessary to detect and quarantine flaky tests.

▶  Scaling an Erlang Cluster to 10,000 Nodes — Maxim Fedorov demonstrates an example of a live Erlang cluster being scaled from just a few nodes to 10,000 machines with no service interruption.

▶  Developing Hundreds of Kubernetes Services at Scale with Airbnb — One of Airbnb’s infrastructure engineers relates Airbnb’s experience with using Kubernetes at scale, focusing on the tooling and developer experience. Note: It’s a talk but there’s a full, readable transcript.

Migrating a Retail Monolith to Microservices — At a recent event, a developer working on a monolithic app maintained by 2 teams explained how they broke things down to 270 microservices managed by 48(!) teams.

Linkerd v2: How Lessons from Production Adoption Resulted in a Rewrite of the Service Mesh — Amongst other things, the popular service mesh moved from being written in Scala to a split between Go and Rust.

Rewriting an API Gateway Service from Clojure to Go — Moving to Go gave AppsFlyer a statically typed language, improved performance, and a more stable solution.

Highly Available MySQL Clustering at WePay Using Orchestrator, Consul and HAProxy — WePay’s engineering team talks about their new highly available MySQL cluster built with HAProxy, Consul and Orchestrator. It improves upon their previous architecture by reducing any downtime from 30 minutes to 40-60 seconds.