Why the future of the cloud will be built on Rust
summary
Linkerd runs a Rust proxy as a lightweight sidecar while its control plane stays in Go, and that Rust proxy is production‑ready and used to avoid classes of failures: compiler‑caught ignored errors, uninitialized values, and concurrency races via the borrow checker and RAII. The stack uses Tokio async runtime, Tower service middleware, Tonic gRPC bindings and kube‑rs for Kubernetes integration, all of which were built or extended for Linkerd and demonstrably work at cloud native scale.
AI-written summary · about
chapters (22)
- 0:00Intro
- 1:02A brief, subjective history of the Cloud
- 1:26The Old World
- 3:25OSS: Config Management
- 4:38Making machines multi-tenant
- 5:38Productizing the Datacenter
- 9:04Decoupling the App from the Host
- 11:26Cloud Native v1
- 13:14Linkerd: A Service Mesh
- 15:31Rust: the right tool at the right time
- 16:33The classical model
- 16:47The modern model
- 17:15The system layers
- 17:44Systems Programming
- 18:51Requiring error handling
- 19:48Preventing runtime errors
- 20:35Opting-in to runtime errors
- 20:57Safe Concurrency
- 22:51Resource Acquisition is Initialization
- 24:47Tokio's Async Ecosystem
- 25:36Modular components
- 29:31Summary
related talks
25mFive Years of Cloud Native Rust.
13mLightning Talk: Writing Service Mesh Control Planes in Rust
25mWTF, Rust for Web Development??
54mRust, Kubernetes, and the Cloud
23mRust in Edge Computing: Enhance Performance & Security
56mThe RUST Language for High‑Load Network Services — Quick Start
27mThe State of Rust Web Frameworks: Pros, Cons & Use Cases
05mRust Here, Rust There, Rust Everywhere! How a Crab Conquers the Cl...