Getting Started

Pre-requisites

  1. Kubernetes cluster

  2. Helm

  3. Access to an existing Prometheus instance in the cluster, or sufficient permissions to deploy a custom instance (preferred).

  4. If using autoscaling, you may need to ask cluster administrators to install CustomResourceDefinitions for KEDA.

Installation

  1. Create a values file with your configuration.

  2. Add Helm repositories for SuperSONIC and its dependencies:

    helm repo add fastml https://fastmachinelearning.org/SuperSONIC
    helm repo add prometheus-community https://prometheus-community.github.io/helm-charts
    helm repo add grafana https://grafana.github.io/helm-charts
    helm repo update
    
  3. Modify the following command to install the chart at your cluster:

    helm install <release-name> fastml/supersonic --values <your-values.yaml> -n <namespace>
    

    Use a unique meaningful lowercase value as <release-name>, for example supersonic-cms-run3. This value will be used as a prefix for all resources created by the chart, unless nameOverride is specified in the values file.

  4. Successfully executed helm install command will print a link to auto-generated Grafana dashboard and other useful information.

Supersonic Grafana dashboard

Uninstall SuperSONIC

helm uninstall <release-name> -n <namespace>

Architecture

The SuperSONIC Helm chart will install components depicted at the diagram below, excluding the model repository, which must be connected by specifying relevant parameters in configuration file (see configuration guide).

SONIC Server Infrastructure