Overview
What is Amazon Redshift: Amazon Redshift is a fully managed, petabyte-scale cloud data warehouse from AWS. Teams use Redshift for near real-time analytics, machine learning, business intelligence, and data lake queries across S3 and third-party sources, plus zero-ETL integrations from Aurora, RDS, and DynamoDB.
How to integrate Amazon Redshift with Datagrid
The Datagrid integration links your Redshift cluster to AI agents that query, transform, and write warehouse data. Use the setup steps below to connect the database, define credentials, and confirm how data sync should run.
Connect your Redshift cluster
Set up the connection in Datagrid:
Open Datagrid and go to Settings > Integrations > Add New.
Select Amazon Redshift from the integration list.
Enter the cluster endpoint, database name, and port (default: 5439).
Provide the administrator username and password.
Test the connection and confirm access to your target schemas.
Save the integration configuration.
Configure authentication
The documented Datagrid setup uses database username and password over a standard connection. Redshift also supports IAM temporary credentials and federated SSO through SAML 2.0 for production environments. See the IAM credentials for the full authentication hierarchy.
Use the same credential fields as protected configuration values:
REDSHIFT_CLUSTER_ENDPOINT=<cluster-endpoint>
REDSHIFT_DATABASE_NAME=<database-name>
REDSHIFT_PORT=5439
REDSHIFT_ADMINISTRATOR_USERNAME=<username>
REDSHIFT_ADMINISTRATOR_PASSWORD=<password>Define data sync
Datagrid reads from Redshift tables, views, and schemas, and writes processed results back. Sync runs bidirectionally. Imports trigger on a defined schedule or in response to source updates. The Redshift object hierarchy is databases, then schemas, then tables and views. View types include regular views, late binding views, and materialized views.
A sync mapping can mirror the Redshift object hierarchy:
sync:
source:
database: <database-name>
schema: <schema-name>
objects:
- <table-or-view-name>
target:
schema: <schema-name>
table: <write-back-table>
direction: bidirectional
trigger:
schedule: <defined-schedule>
source_updates: trueAfter connection, authentication, and sync rules are saved, Datagrid agents can execute warehouse workflows against Redshift data and return processed outputs to Redshift.
Why use Amazon Redshift with Datagrid
You can use the integration when warehouse data needs to become completed work that teams can act on. Here are some examples:
Autonomous warehouse queries: Agents run SQL against Redshift tables and return answers without you writing queries or exporting data.
Write-back transformations: Agents execute transformations and load results back to Redshift, so processed data lands where reporting tools already read it.
Schedule and event triggers: Agents import data on a fixed schedule or when source tables update, so agent context stays current.
AWS Data API option: The Redshift Data API provides a secure HTTP endpoint with asynchronous calls and no persistent connections to maintain for AWS workflows around Redshift.
Agentic AI on warehouse data: Pair Redshift with agents to classify customers, summarize records, and analyze sentiment against structured data at scale.
Cross-system routing: Agents move outputs from Redshift into the operational tools your project teams use daily, so analysis turns into action.
What you can build with Amazon Redshift Datagrid integration
Datagrid's AI agents can use Redshift as the warehouse record for reporting, enrichment, cross-system routing, and exception detection workflows. Use the following patterns when warehouse records need to trigger completed work across project systems:
Automated reporting pipelines: Build an agent that queries daily transactional tables, calculates metrics, and writes summary tables back to Redshift for dashboards.
Data enrichment workflows: Build an agent that reads raw records from Redshift, enriches them with classifications or summaries, and loads the results into a new table.
Cross-platform data sync: Build an agent that extracts query results from Redshift and routes them into CRM, document, or communication tools.
Exception detection on warehouse data: Build an agent that scans Redshift tables on a schedule, detects anomalies, and alerts your team.
Resources and documentation
Data API: Plan secure HTTP endpoint access, asynchronous calls, and supported statements.
IAM credentials: Review authentication methods for external connections.
Configuring connections: Review JDBC, Python, and ODBC connection behavior.
Frequently asked questions
What authentication does the Datagrid Redshift integration use?
The documented Datagrid setup uses a database administrator username and password with the cluster endpoint, database name, and port 5439. Redshift also supports IAM temporary credentials and federated SSO via SAML 2.0 for production use. Review the options for providing IAM credentials to plan the right method.
Can Datagrid agents write data back to Redshift?
Yes. Agents ingest data from Redshift tables, execute transformations, and write results back. Separately, the Redshift Data API supports DML, DDL, COPY, and UNLOAD statements through a secure HTTP endpoint for AWS workflows around Redshift.
How often does Datagrid sync data from Redshift?
Imports trigger on a defined schedule or in response to source updates. For continuous replication into Redshift itself, AWS zero-ETL integrations provide near real-time, change-data-capture pipelines from Aurora, RDS, and DynamoDB without custom code.
Does the Redshift Data API require a persistent connection?
The Data API uses a secure HTTP endpoint and AWS SDK integrations, so you do not maintain a persistent database connection. Calls are asynchronous. Review the Data API documentation for endpoint and SDK details.
What data objects can agents query in Redshift?
Agents work against the Redshift object hierarchy of databases, schemas, and tables or views. View types include regular views, late binding views, and materialized views, as listed in SHOW TABLES. Column metadata, including data types and sort keys, is available through SHOW COLUMNS.
Similar integrations
Snowflake: Alternative cloud data warehouse for cross-platform comparisons, migrations, and hybrid analytics alongside Redshift.
BigQuery: Google Cloud's serverless data warehouse often compared or used alongside Redshift for multi-cloud analytics and query federation.
Databricks: Lakehouse platform for ETL, streaming, and advanced ML that complements Redshift's warehousing with unified data engineering workflows.
Amazon AWS S3: Core object storage for COPY/UNLOAD operations and lakehouse storage used to ingest and export Redshift data efficiently.
Amazon RDS: Transactional database source for zero-ETL replication and federated queries into Redshift for near real-time analytics.
Amazon Aurora: Managed relational database frequently used with Redshift's zero-ETL pipelines to stream transactional data for analytics.