Migrate to Polaris from Rockset
This guide walks through the process to seamlessly migrate your data to Imply Polaris on AWS from Rockset.
The following diagram demonstrates the end-to-end process of the migration to Polaris. Shaded boxes represent steps taken within Polaris, and unshaded boxes represent steps taken outside Polaris, either in AWS or Rockset.
Sign up for Polaris
If you haven't already, sign up for a Polaris trial. You need a Polaris organization to host your Rockset data.
Optionally, run through the Polaris quicksart to learn about Polaris's features.
Create a staging S3 bucket
Launch the AWS Management Console and create an S3 bucket to stage your data. To ensure you assign the correct permissions to your bucket, refer to Rockset's example IAM policies.
Export your data to S3
Login to Rockset and use the INSERT INTO s3
function to export your data to the staging bucket you created.
For best results export your data in Parquet format.
Import your data into Polaris
Follow the steps in Connect to Amazon S3 to create a connection from Polaris to S3. Use the example IAM permissions policy to ensure you assign the correct permissions for Polaris.
Refer to the Guide for Amazon S3 ingestion for detailed instructions.
The Imply team can help with ingestion if you need it.
Connect your client
You can run queries against your data in Polaris using the SQL workbench.
When you're ready to connect your client applications, you can use the following APIs:
You can also connect a number of third-party business intelligence tools.
Migrate your queries
Rewrite your Rockset queries using Druid SQL. Then submit them to Polaris using the Query API.
If you need assistance, the Imply team will help you migrate your queries.
Learn more
See the following topics for more information:
- Connect to S3 for information on creating an Amazon S3 connection in the Polaris UI.
- Ingest data from S3 by API for information on using the Connections v1 API and the Jobs v1 API to ingest data from Amazon S3.