Arize AI
Search…
Installation
Installation Details for Arize On-Prem Deployment

Overview

The installation requires the following TAR file that will be supplied by the Arize team. The TAR file includes all manifests, images and scripts to install the Arize platform.
arize-images-<target>-<hash>.tar
- deployment
|-base
|-azure
|-gcs
|-s3
|-db
deploy.sh

1. Pre-Deployment

The Arize team can help size the cluster based on team requirements.
  • Storage bucket entities need to be created for Arize
  • A service account needs to be created with access to the bucket storage and Kubernetes cluster
  • IP address and VPC setup should be discussed with the Arize team. Our team can help pre-configure the files for network setup based on required deployment options.

2. Deployment

The Arize team will provide a cloud or on-premise specific tarball that includes images, config and install script.
To get started quickly, you can use the quick start manifest which will install the Arize AI Platform as well as some commonly used components:
tar -zxvf arize-distribution-v-12.3.tar.gz
kustomization.yaml
apiVersion: kustomize.config.k8s.io/v1beta1
kind: Kustomization
namespace: arize
resources:
- manifests/...
images:
- name: ...
secretGenerator:
- name: ...
literals:
- KEYS= ...
type: Opaque
quickstart.sh is the main install script. This uses kubectl to apply all resources to your cluster, then runs initial application and database setup.
After installation, endpoints for sending data from the SDK and for accessing the Platform UI are available for consumption by other applications running in the cluster. These endpoints can be exposed to infrastructure outside of kubernetes through additional Ingress configuration.
Initial default login is based on the default login and password in the configuration setup. On initial login the admin interface will clearly highlight platform status and go indicator for initial use.
The deploy script is based on 10 flags used to deploy the distribution.
deploy.sh command
#Help CLI
deploy.sh -h
Usage: ./distribution/deploy.sh
[ -k CLUSTER ]
[ -c CLOUD_PROVIDER ]
[ -a SERVICE_ACCOUNT_PATH ]
[ -x AWS_CREDENTIALS_PATH ]
[ -p AWS_CONFIG_PATH ]
[ -z AWS_REGION ]
[ -g GAZETTE_BUCKET ]
[ -d DR_BUCKET ]
[ -s POSTGRES_PASSWORD ]
[ -e SMTP_PASSWORD ]
[ -o SMTP_HOST ]
[ -w SMTP_USER ]
[ -n SMTP_SENDER_EMAIL ]
[ -r IMAGE_REGISTRY_URI ]
[ -l PRELOADED_IMAGES_URI ]
[ -t PRELOADED_IMAGES_TAG ]
[ -y MANIFEST_DST ]
[ -i Use this flag to indicate script should configure Arize services only in in an already deployed cluster (e.g apply secrets, initialize postgres, configure gazette) ]
  1. 1.
    CLOUD_PROVIER: gcp/aws/azure
  2. 2.
    CLUSTER: The cluster name on kubeconfig of the deployment
  3. 3.
    SERVICE_ACCOUNT_PATH: (GCP only)The path to the service account credentials for gcp installs
  4. 4.
    AWS_CREDENTIALS_PATH: (AWS only)The absolute path to aws credential file (only single credential supported)
  5. 5.
    AWS_CONFIG_PATH: (AWS only) The absolute path to aws config file
  6. 6.
    AWS_REGION: (AWS only) Cluster region
  7. 7.
    GAZETTE_BUCKET: The bucket name to hold gazette events
  8. 8.
    DR_BUCKET: The storage bucket to hold ui data
  9. 9.
    POSTGRES_PASSWORD: The postgres db admin password
  10. 10.
    SMTP_PASSWORD: The password for the SMTP service
  11. 11.
    SMTP_USER: The user for the SMTP service
  12. 12.
    SMTP_HOST: The host endpoint for the SMTP service
  13. 13.
    SMTP_SENDER_EMAIL: The smtp authenticated address emails should come from. e.g. From: [email protected]
  14. 14.
    IMAGE_REGISTERY_URI: The registry of the images for install
  15. 15.
    PRELOADED_IMAGES_PATH: The path to the preloaded image files
  16. 16.
    PRELOADED_IMAGES_TAG: The tags of the preloaded images
Running the script executes a number steps that include:
  • Applying the secrets
  • Applying the manifests
  • Preparing the Database
  • Starting the consumer applications
  • Finally starting the User Interface and SDK receiver
Output of the script will look as follows:
Output of Deploy.sh

3. Post Deploy

After deployment teams should confirm:
  • Secrets have been applied
  • All Arize Kubernetes services are green and up
Test that the User Interface is live by accessing at localhost:4040:
The Arize team will typically work on completing the installation through help in setting up IP addresses, initial login accounts and testing the end to end system.
Questions? Email us at [email protected] or Slack us in the #arize-support channel