Why is Duco migrating from Impala to Redshift?
It became obvious to us that for improving our service we need to adjust to more modern database architecture. Redshift is an analytical column-oriented database, fit for purpose for large-scale analytical use cases. Upgrading our architecture also unlocks exciting new capabilities in the future.
Redshift provides significant performance improvements:
- Data sync: For 1–2 million row runs, sync times dropped from up to 50 minutes to ~1 minute.
- Query execution: Up to 10x faster, enabling rapid data exploration and reporting.
How can I get connected?
That depends on what tool you’ll be using with the Data Platform.
- Most BI tools (e.g., Power BI, Tableau, Looker) have a native Redshift connector.
- You might be required to install the Redshift driver in your BI tool, please check with your BI administrator.
- You can also connect via ODBC/JDBC drivers.
- Please ensure TCP port 5439 is open in your network.
If you need assistance, our Client Support support@du.co team can guide you through setup.
Is there a UI for the Data Platform?
No. We provide direct database access. You have full flexibility to use any tool of your choice (BI tools, notebooks, ETL platforms) to connect to the data platform.
What’s the authentication method?
Authentication is currently via username and password.
- If you are migrating from the legacy platform, you will retain your current password.
- If you are new to the Data Platform, our Client Success team will guide you through obtaining credentials.
Will I be able to connect from my BI tool?
Yes. Most modern BI tools support native Redshift connectivity as well as ODBC/JDBC drivers. Please check with your BI tool administrator if a driver installation or connector configuration is required.
Which port and protocol does Redshift use?
- Protocol: PostgreSQL wire protocol
- Port: 5439 (default)
Please ensure this is open and allowed by your internal network/security team.
How is security managed?
- All connections use SSL encryption in transit.
- Authentication is handled via username/password.
- IP allowlisting is required
- Access is controlled at the database and schema level.
- We recommend that you manage access to credentials within your organization in accordance with your internal security policies.
Can multiple users/teams connect at the same time?
Yes. Redshift supports concurrent connections.
Do I need to add date filters to every query?
No. In Impala, time-partitioned views were required for performance. In Redshift, date filtering is optional. However, it's still good practice for large datasets.
How do I find my custom field names?
Run SELECT match_fields FROM results_custom_fields LIMIT 10; to see the full JSON and discover the exact field names (they are case sensitive).
What schema do I use?
All tables are in {tenant}_dataplatform_v1. Either prefix your tables or set the search path.
Is AWS PrivateLink available?
Yes. For clients requiring private, non-internet-routed connectivity, AWS PrivateLink is available. Contact the Client Success team for setup.