Linkerd 2.0 is now generally available with a new service sidecar design

Linkerd 2.0 is now generally available. Linkerd is a transparent proxy that adds service discovery, routing, failure handling, and visibility to modern software applications. Linkerd 2.0 brings two significant changes. First, Linkerd 2.0 is completely rewritten to to be faster and smaller than Linkerd 1.x. Second, Linkerd moves beyond the service mesh model to running on a single service. It also comes with a focus on minimal configuration, a modular control plane design, and UNIX-style CLI tools.

Let’s understand what each of these changes mean.

Smaller and Faster

Linkerd has undergone a complete change to become faster and smaller than its predecessor. Linkerd 2.0’s data plane is comprised of ultralight Rust proxies which consume around 10mb of RSS and have a p99 latency of <1ms. Linkerd’s minimalist control plane (written in Go) is similarly designed for speed and low resource footprint.

Service sidecar design

It also adopts a modern service sidecar design from the traditional service mesh model. The traditional service mesh model has two major problems. First, they add a significant layer of complexity to the tech stack. Second they are designed to meet the needs of platform owners undermining the service owners.

Linkerd 2.0’s service sidecar design offers a solution to both. It allows platform owners to build out a service mesh incrementally, one service at a time, to provide security and reliability that a full service mesh provides. More importantly, Linkerd 2.0 addresses the needs of service owners directly with its service sidecar model to its focus on diagnostics and debugging.

Linkerd 2.0 at its core is a service sidecar, running on a single service without requiring cluster-wide installation. Even without having a whole Kubernetes cluster, developers can run Linkerd and get: