AI-Driven Data Visualization with AWS QuickSight

Question

Pulumi · Accepted Answer

To achieve AI-driven data visualization with AWS QuickSight using Pulumi in Python, we'll follow these steps:

Set up the AWS provider: Ensure that your AWS credentials are properly configured to allow Pulumi to manage resources on your behalf.
Create a QuickSight Dataset: You need a dataset which QuickSight will use to create visualizations. This dataset can be created within QuickSight, or you can connect QuickSight to an existing data source.
Create a QuickSight Theme: A theme controls the overall look and feel of your QuickSight dashboard, including colors and style.
Create a QuickSight Analysis: An analysis in QuickSight is a workspace where you can create visualizations, perform exploratory data analysis, and arrange those visualizations how you like.
Create a QuickSight Dashboard: Finally, you create a dashboard from your analysis. Dashboards are read-only snapshots of your analysis that you can share with others in your organization.

Below is a basic Pulumi program in Python that sets up a QuickSight dashboard. Note that this example assumes you have the necessary permissions and that your data source is already set up in AWS. The Pulumi AWS provider, specifically the pulumi_aws package, is used to interact with QuickSight resources.

Important: Be sure the Pulumi CLI and AWS CLI are installed, you're logged in to AWS, and the Pulumi stack is properly initialized before running this program.

import pulumi
import pulumi_aws as aws

# Assume AWS configuration and credentials are already set up.

# Create a QuickSight dataset.
# Replace these values with the actual references to your data source
data_source_arn = "<data-source-arn>"  # This ARN is the resource ID for your data source
data_set = aws.quicksight.DataSet("my-data-set",
                                  aws_account_id="<aws-account-id>",
                                  data_source_arn=data_source_arn,
                                  import_mode="DIRECT_QUERY")  # This can also be SPICE for importing the data.

# Alternatively, you can define source entities for your DataSet, such as S3 sources, SQL queries, etc.

# Create a QuickSight theme for styling your visuals.
theme = aws.quicksight.Theme("my-theme",
                             aws_account_id="<aws-account-id>",
                             base_theme_id="DEFAULT",  # You can customize this further as needed
                             name="CustomTheme",
                             permissions=[{
                                 "actions": ["quicksight:DescribeTheme"],
                                 "principal": "*",
                             }])

# Create a QuickSight analysis.
analysis = aws.quicksight.Analysis("my-analysis",
                                   analysis_id="my-analysis-id",
                                   aws_account_id="<aws-account-id>",
                                   source_entity={
                                       "sourceTemplate": {
                                           "arn": "<template-arn>",
                                           "dataSetReferences": [{
                                               "dataSetArn": data_set.arn,
                                               "dataSetPlaceholder": "DataSetPlaceholder",
                                           }],
                                       },
                                   })

# Create a QuickSight dashboard based on the analysis.
dashboard = aws.quicksight.Dashboard("my-dashboard",
                                     dashboard_id="my-dashboard-id",
                                     aws_account_id="<aws-account-id>",
                                     permissions=[{
                                         "actions": ["quicksight:DescribeDashboard",
                                                     "quicksight:ListDashboardVersions",
                                                     "quicksight:UpdateDashboardPermissions",
                                                     "quicksight:QueryDashboard",
                                                     "quicksight:UpdateDashboard",
                                                     "quicksight:DeleteDashboard",
                                                     "quicksight:DescribeDashboardPermissions",
                                                     "quicksight:UpdateDashboardPublishedVersion"],
                                         "principal": "*",
                                     }],
                                     source_entity={
                                         "sourceAnalysis": {
                                             "arn": analysis.arn,
                                             "dataSetReferences": [{
                                                 "dataSetPlaceholder": "DataSetPlaceholder",
                                                 "dataSetArn": data_set.arn,
                                             }],
                                         },
                                     },
                                     version_description="1")

# Export the URL of the dashboard.
pulumi.export("dashboard_url", dashboard.url)

In this program:

We're creating a QuickSight dataset that uses an existing data source. Replace "<data-source-arn>" and "<aws-account-id>" with appropriate values for your AWS environment.
A QuickSight theme named CustomTheme is created, which you can customize further as needed.
An analysis is set up with a given analysis_id, pointing to a template ARN, which is necessary if you want to create dashboards or analyses programmatically. The sourceTemplate indicates the template you've based your dataset on and where the placeholder in the template is related to the dataset.
A dashboard is then created from the analysis with a set of permissions allowing various QuickSight actions.
Finally, the URL of the dashboard is exported so that it can be accessed outside of Pulumi.

Remember to replace the placeholder ARNs, IDs, and account information with the actual values from your AWS environment.

Please consult the AWS QuickSight documentation to dive deeper into the specifics of creating themes, datasets, and so forth, as these will depend on your specific data source and visualization requirements.