Kubernetes release optimizer built for developers, DevOps and MLOps engineers, SREs, and data scientists.
Iter8 experiments make it simple to collect performance and business metrics for apps and ML models, assess, compare and validate multiple app/ML model versions, promote the winning version, and maximize business value in each release.
Experiment charts are Helm charts with a special structure that contain reusable experiment templates. Iter8 combines experiment charts with user supplied values to generate runnable
experiment.yaml files. Iter8 experiment charts enable you to launch powerful release optimization experiments in seconds. Their usage is described in depth in various Iter8 tutorials.
Iter8 hub refers to a Helm repo that hosts Iter8 experiment charts. The official Iter8 hub is located at https://iter8-tools.github.io/hub/. You can create, package and host Iter8 experiment charts in any Helm repo and use them with Iter8 CLI.
Features at a glance¶
Load testing with SLOs
Iter8 experiments can generate requests for HTTP and gRPC services, collect built-in latency and error-related metrics, and validate SLOs.
Grow your business with every release. Iter8 experiments can compare multiple versions based on business value and identify a winner.
Simple to use
Get started with Iter8 in seconds using pre-packaged experiment charts. Run Iter8 experiments locally, inside Kubernetes, or inside your CI/CD/GitOps pipelines.
Use with any app, serverless, or ML framework. Iter8 works with Kubernetes deployments, statefulsets, Knative services, KServe/Seldon ML deployments, or other custom Kubernetes resource types.
Iter8 is written in
go and builds on a few awesome open source projects including: