Knative 1.0: Serverless on Kubernetes Reaches GA

Introduction

Knative 1.0 — Serverless on Kubernetes Reaches GA — was released on November 16, 2021.

This is a practical update aimed at making day‑to‑day Kubernetes work a bit more predictable.

In this release: Knative 1.0 achieves General Availability, delivering production-ready serverless workloads with stable APIs, enhanced autoscaling, and mature eventing capabilities.

Serving v1 APIs Stable

• Service, Configuration, and Route resources graduate to v1 API version, providing long-term stability guarantees for production deployments.
• Revision management gains enhanced lifecycle controls, enabling precise traffic splitting and gradual rollouts.
• Domain mapping supports custom domains and TLS termination, simplifying multi-tenant deployments without external ingress controllers.
• Container concurrency and scale bounds expose fine-grained controls for optimizing resource utilization and latency.

Eventing Production Ready

1. Broker v1beta1 standardizes event delivery with configurable retries, dead-letter queues, and tracing integration.
2. Trigger filtering enables attribute-based event routing, supporting complex event-driven architectures.
3. Source ecosystem matures with production-ready connectors for Kafka, RabbitMQ, GitHub, and cloud event sources.
4. Channel implementations provide reliable event delivery with pluggable backends (InMemory, Kafka, NATS).

Autoscaling Excellence

• KPA (Knative Pod Autoscaler) delivers sub-second scale-to-zero and rapid scale-up, handling traffic bursts with minimal cold starts.
• Activator optimizations reduce queueing latency and improve request routing during scale transitions.
• Concurrency-based scaling automatically adjusts replica counts based on request load, optimizing resource usage.
• Burst capacity settings prevent traffic spikes from overwhelming newly scaled pods.

Operational Maturity

• Health checks and readiness probes ensure traffic only routes to healthy revisions.
• Metrics integration with Prometheus provides visibility into request rates, latencies, and autoscaling decisions.
• Distributed tracing support via OpenTelemetry enables end-to-end request observability.
• Multi-cluster support through service mesh integration (Istio, Linkerd) enables global serverless deployments.

Getting Started

kubectl apply -f https://github.com/knative/serving/releases/download/v1.0.0/serving-crds.yaml
kubectl apply -f https://github.com/knative/serving/releases/download/v1.0.0/serving-core.yaml
kubectl apply -f https://github.com/knative/net-istio/releases/download/v1.0.0/release.yaml

Deploy a serverless service:

kn service create hello \
  --image gcr.io/knative-samples/helloworld-go:latest \
  --env TARGET="Knative 1.0"

Summary

Knative 1.0 marks the beginning of a new era for serverless on Kubernetes, providing teams with a mature, scalable platform for building event-driven applications with confidence.