Setup p0 on your on-prem server on Google Cloud Provider (GCP)

Welcome to the on-premise installation instructions for p0 on Google Cloud Provider (GCP). Follow this guide to get set up with your own version of p0.

  1. Ensure that you have gcloud [command line interface for Google Cloud], which is authenticated. And configured with the project ID you want to use. You can refer Managment through CLI documentation for more information.

  2. Now, create a service account. You can create a service account either using the CLI (see command below) or using web console.

    • Note #1: Make sure that this service account has access to Vertex API for accessing the Claude 3 Haiku model. Please refer to this page to understand how to configure the model.

    • Note #2: For new gCloud accounts, Google gives you a few free cloud credits but you’ll need to create a billing account in order to get Claude 3 Haiku model.

    • If you're configuring the service via CLI, you can use the command below to do it. In the command below, the p0-vertex-ai-sa is the name of the service account:

gcloud iam service-accounts create p0-vertex-ai-sa

gcloud iam service-accounts create p0-vertex-ai-sa

gcloud iam service-accounts create p0-vertex-ai-sa

  1. Provide required roles to the newly created service account, replace <PROJECT_ID> in the same.

gcloud projects add-iam-policy-binding <PROJECT_ID> --member="serviceAccount:p0-vertex-ai-sa@<PROJECT_ID>.iam.gserviceaccount.com" --role=roles/aiplatform.user

gcloud projects add-iam-policy-binding <PROJECT_ID> --member="serviceAccount:p0-vertex-ai-sa@<PROJECT_ID>.iam.gserviceaccount.com" --role=roles/aiplatform.user

gcloud projects add-iam-policy-binding <PROJECT_ID> --member="serviceAccount:p0-vertex-ai-sa@<PROJECT_ID>.iam.gserviceaccount.com" --role=roles/aiplatform.user

  1. You can now use gcloud CLI to import the p0-visiblity-dashboard image using below command (when prompted allow all permissions). Note that the process may take upto 15 mins to complete. Please be patient.

    • In case of any error (network failure, Google failing to create bucket, etc), retry the command.

gcloud compute images import imported-p0-vd-image --source-file gs://public-p0/p0-visibility-dashboard-ami.vmdk --os ubuntu-2204

gcloud compute images import imported-p0-vd-image --source-file gs://public-p0/p0-visibility-dashboard-ami.vmdk --os ubuntu-2204

gcloud compute images import imported-p0-vd-image --source-file gs://public-p0/p0-visibility-dashboard-ami.vmdk --os ubuntu-2204

  1. Once it completes successfully, you will see message [import-image]: 2024-05-23T06:44:25Z Finished making disk bootable in your termnial.

  2. Now, go to your browser, login into GCP and Go to Compute Engine > Storage > Images [Direct link]

  3. You should see the imported image, if not, search for the image name: imported-p0-vd-image using filter.

  4. Once you locate the image, click on 3-dot Action Column next to image, then click on Create Instance

  1. Provide the name as “p0-visibility-dashboard”, Select the desired configuration. Select the machine type as e2-standard-4 [4 vCores and 16 GB RAM or equivalent]. Please do not select Spot Instances as they can get terminated during the runs.

  1. Select the service account created previously in the dropdown

  1. Select “Allow HTTP Traffic” and “Allow HTTPS Traffic” option in the VM Creation page.

  1. Click on “Create” button at the bottom of the page.

  1. In about 5 mins, You should visit the external IP of the newly created instance and start with the onboarding.

Contact us

If you are facing any trouble setting up your on-prem application, reach out to us at contact[at]p0[dot]inc

Contact us

If you are facing any trouble setting up your on-prem application, reach out to us at contact[at]p0[dot]inc

Contact us

If you are facing any trouble setting up your on-prem application, reach out to us at contact[at]p0[dot]inc

© 2024 p

0

. All rights reserved.

© 2024 p

0

. All rights reserved.

© 2024 p

0

. All rights reserved.