connector

Amazon Athena

Serverless SQL over data in Amazon S3.

Connection fields

Field	Required	Description
`aws_access_key`	Yes	AWS access key ID.
`aws_secret_key`	Yes	AWS secret access key.
`region`	Yes	AWS region, e.g. "us-east-1".
`s3_output_location`	Yes	S3 URI Athena writes query results to, e.g. "s3://bucket/prefix/".
`work_group`	No	Athena workgroup to run under.
`catalog`	No	Leave blank to use Athena's default AwsDataCatalog. Set this only for a custom Athena data catalog.
`database`	Yes	Database/schema to query.

Uses the static access-key/secret you configure here — never the ambient AWS credential chain (environment variables, instance profile) of the machine arcodash runs on.

Example configuration

{
  "aws_access_key": "AKIA••••••••",
  "aws_secret_key": "••••••••",
  "region": "us-east-1",
  "s3_output_location": "s3://my-athena-results/arcodash/",
  "work_group": "primary",
  "database": "analytics"
}

Caveats

Not yet live-verified against real AWS or a paid LocalStack Pro instance (community LocalStack doesn't implement Athena) — type mapping and pagination are unit-tested against constructed SDK responses, not a live query.
Every query has to finish writing its result set to s3_output_location before arcodash can read it back, so Athena queries pick up S3's latency on top of the query's own runtime.