Overview and State of Linkerd
summary
Linkerd presents itself as a CNCF‑graduated, production‑proven service mesh running four plus years in production and cited for securing 22k pods in a case study. It ships a Rust-based ultra-light proxy, compiles for ARM, adds a multi‑core runtime and makes mTLS for TCP by default. Recent releases add server‑side authorization, grpc retries, composable extensions (viz, tracing, multi‑cluster), and a multi‑cluster failover controller; roadmap targets client policies (JWT, circuit breaking, header routing) and mesh expansion beyond Kubernetes.
AI-written summary · about
chapters (28)
- 0:00Start
- 2:20Service Meshes
- 4:42Rust
- 6:48Approach to Security
- 7:39How Does Linguidi Compare to Istio
- 10:16Multi-Multi-Core Proxy Runtime
- 10:31Prometheus Optional
- 13:02Authorization Policies
- 13:20Grpc Retries
- 14:32Architecture
- 15:09Policy Controller
- 15:29Certificate Management
- 15:53Extensions
- 16:38Tap Injector
- 17:13What's Next for Linguity
- 18:50The Cni Plug-In
- 19:08Failover Controller for Multi-Cluster
- 19:35Server Authorizations
- 19:58Policy Attachment
- 22:30Community
- 23:11Third Party Security Audits and Benchmarks
- 23:21Hero Program
- 23:43Anchor Program
- 25:32Edge Releases
- 27:44Proxy Injection
- 31:04Ebpf Mode of Running a Service Mesh
- 32:34Ebpf
- 33:44Ebpf Service Mesh
related talks
28mOverview and State of Linkerd
35mDeep Dive: Linkerd
35mIntro: Linkerd
06mKeynote: Linkerd Project Update
32mLinkerd and the Service Mesh
30mLinkerd Update: Gateway API, Client-Specific Policy, Federated Services, Multicluster, Rust, & More!
12mService Mesh & Kubernetes Multicluster Communication With Linkerd
01h 10mService Mesh Introduction