Amazon Redshift Driver v1.9.0¶

Driver Version v1.9.0 Release Date 2026-11-10 Tested With Amazon Redshift aws

This driver provides access to Amazon Redshift (commonly referred to as just “Redshift”).

Note

This project is not affiliated with Amazon.

Installation¶

The Redshift driver can be installed with dbc :

dbc install redshift

Connecting¶

To connect, edit the uri, redshift.cluster_type, redshift.workgroup_name, and redshift.db_name options below to match your environment and run the following:

from adbc_driver_manager import dbapi

dbapi.connect(
    driver="redshift",
    db_kwargs={
        "uri": "postgresql://localhost:5439",
        # "uri": "postgresql://<cluster hostname>:<cluster port>", # for direct connection
        "redshift.cluster_type": "redshift-serverless",  # for Redshift Serverless
        # "redshift.cluster_type": "redshift-iam", # for Redshift Provisioned with IAM auth
        # "redshift.cluster_type": "redshift", # for Redshift Provisioned with user/password auth
        "redshift.workgroup_name": "<WORKGROUP_NAME>",  # for Redshift Serverless
        # "redshift.cluster_identifier": "<CLUSTER IDENTIFIER>", # for Redshift Provisioned
        "redshift.db_name": "sample_data_dev",
    },
)

Note: The example above is for Python using the adbc-driver-manager package but the process will be similar for other driver managers.

Connection String Format¶

Connection strings are passed with the uri option. Redshift’s URI syntax supports two primary forms:

Explicit endpoint, for direct connections or standard user/password authentication:
```
redshift://[user[:password]@]host[:port]/dbname[?param1=value1&param2=value2]
```
Endpoint discovery, for IAM authentication and serverless connections:
```
redshift:///dbname[?param1=value1&param2=value2]
```
Note the triple slash (///). This indicates an empty host, signaling the driver to discover the endpoint using the provided query parameters (e.g., cluster_identifier or workgroup_name).

Components:

Scheme: redshift:// (required)
User/Password: Optional (for standard user/password authentication with cluster_type=redshift)
- Ignored if cluster_type is redshift-serverless
- If cluster_type is redshift-iam, then if omitted, the driver will log in with the current IAM identity; else it will fetch credentials for the given username
- For cluster_type redshift, the driver will fetch credentials for the given username
Host/Port: The Redshift cluster endpoint (e.g., my-cluster.c…com:5439)
- If omitted (using redshift:///), the driver will discover this using the AWS SDK
Database: Required (e.g., /dev)
Query Parameters: Configuration options

See Amazon Redshift JDBC Connection URL for connection URL format and Amazon Redshift JDBC Configuration Options for complete parameter reference.

Note

Reserved characters in URI elements must be URI-encoded. For example, @ becomes %40.

Examples:

redshift://admin_user:secret@my-cluster.region.redshift.amazonaws.com:5439/dev
redshift:///dev?cluster_type=redshift-iam&cluster_identifier=my-cluster
redshift:///dev?cluster_type=redshift-serverless&workgroup_name=my-workgroup
redshift://admin:pass@localhost:5439/dev?sslmode=disable
redshift:///dev?cluster_type=redshift-serverless&workgroup_name=my-workgroup&aws_region=us-west-1&aws_access_key_id=AKIA…&aws_secret_access_key=secret
redshift:///dev?cluster_type=redshift-iam&cluster_identifier=my-cluster&auth_provider=BrowserIdcAuthPlugin&idc_region=us-east-1&issuer_url=https://example.com
redshift:///dev?cluster_type=redshift&cluster_identifier=my-cluster&username=newuser&auto_create_user=true

The driver also supports standard PostgreSQL connection strings (e.g., postgresql://user:pass@host:port/dbname) as the uri option. When using this format, all Redshift-specific parameters (e.g., redshift.cluster_type, redshift.workgroup_name) must be passed as separate connection options. However, using the standard redshift:// URI is recommended.

Feature & Type Support¶

Feature		Amazon Redshift
Bind Parameters		✅
Bulk Ingestion	Create	✅
	Append	✅
	Create/Append	✅
	Replace	✅
	Temporary Table	✅
	Target Catalog	✅
	Target Schema	✅
	Non-nullable fields are marked NOT NULL	✅
Catalog (GetObjects)	depth=catalogs	✅
	depth=db_schemas	✅
	depth=tables	✅
	depth=columns (all)	✅
Get Parameter Schema		✅
Get Table Schema		✅
Prepared Statements		✅
Transactions		✅

Types¶

Database to Arrow¶

Database Type	Amazon Redshift
BIGINT	int64
BOOLEAN	bool
DATE	date32[day]
DECIMAL	decimal128
DOUBLE PRECISION	double
GEOGRAPHY	extension<geoarrow.wkb>
GEOMETRY	extension<geoarrow.wkb>
HLLSKETCH	extension<arrow.opaque[storage_type=string, type_name=HLLSKETCH, vendor_name=Amazon Redshift]>
INTEGER	int32
INTERVAL DAY TO SECOND	month_day_nano_interval
INTERVAL YEAR TO MONTH	month_day_nano_interval
REAL	float
SMALLINT	int16
SUPER	extension<arrow.json>
TIME	time64[us]
TIMESTAMP	timestamp[us]
TIMESTAMP WITH TIME ZONE	timestamp[us] (with time zone)
TIMETZ	extension<arrow.opaque[storage_type=string, type_name=TIMETZ, vendor_name=Amazon Redshift]> [1]
VARBYTE	binary
VARCHAR	string

Arrow to Database¶

Arrow Type	Amazon Redshift Type
	Bind	Ingest
	binary	VARBYTE ⚠️ [2]	VARBYTE
binary_view	VARBYTE ⚠️ [2]	❌ [9]
bool	BOOLEAN
date32[day]	DATE
decimal128	DECIMAL	DECIMAL, NUMERIC ⚠️ [10]
double	DOUBLE PRECISION
extension<arrow.json>	SUPER	SUPER ⚠️ [11]
extension<geoarrow.wkb>	GEOGRAPHY, GEOMETRY ⚠️ [3] [4]	(NA/not tested)
extension<geoarrow.wkt>	GEOGRAPHY, GEOMETRY ⚠️ [5] [6]	GEOMETRY
fixed_size_binary	VARBYTE ⚠️ [2]	❌ [12]
float	REAL ⚠️ [7]	REAL
halffloat	REAL	(NA/not tested)
int16	SMALLINT
int32	INTEGER
int64	BIGINT
large_binary	VARBYTE ⚠️ [2]	VARBYTE
large_string	VARCHAR	VARCHAR(MAX)
list	(NA/not tested)	SUPER
string	VARCHAR	VARCHAR(MAX)
string_view	VARCHAR	❌ [13]
struct	SUPER
time32[ms]	TIME	❌ [14]
time32[s]	TIME	❌ [14]
time64[ns]	TIME	❌ [14]
time64[us]	TIME	❌ [14]
timestamp[ms]	TIMESTAMP
timestamp[ms] (with time zone)	TIMESTAMP(3) WITH TIME ZONE	TIMESTAMPTZ
timestamp[ns]	TIMESTAMP [8]	TIMESTAMP
timestamp[ns] (with time zone)	TIMESTAMP(9) WITH TIME ZONE	TIMESTAMPTZ
timestamp[s]	TIMESTAMP
timestamp[s] (with time zone)	TIMESTAMP(0) WITH TIME ZONE	TIMESTAMPTZ
timestamp[us]	TIMESTAMP
timestamp[us] (with time zone)	TIMESTAMP(6) WITH TIME ZONE	TIMESTAMPTZ

Options¶

Connection Options¶

redshift.application_name

Type: string.

The application name to present to the server.

redshift.auth.idc_client_display_name

Type: string.

For browser-based authentication, the client name to display in the AWS IAM Identity Center.

redshift.auth.idc_region

Type: string.

For browser-based authentication, the AWS region of the IdC.

redshift.auth.idp_response_timeout_seconds

Type: integer. Default: 120

For IdP token authentication, how long to wait (in seconds) for the SSO flow to complete.

redshift.auth.issuer_url

Type: string.

For browser-based authentication, the issuer URL.

redshift.auth.listen_port

Type: integer. Default: 7890.

For browser-based authentication, the local port to listen on. Use port 0 to have the operating system assign a free port.

redshift.auth.provider

Values: userpass, BrowserIdcAuthPlugin, IdpTokenAuthPlugin. Default: userpass.

How to authenticate. The default is user-password; BrowserIdcAuthPlugin triggers an authentication flow in the browser, while IdpTokenAuthPlugin fetches a token from an IdP.

redshift.auth.token

Type: string.

An auth token to use with IdP token authentication.

redshift.auth.token_endpoint.request_data

Type: string.

The form-urlencoded request body to POST to the IdP token endpoint.

redshift.auth.token_endpoint.request_headers

Type: string.

Request headers to send to the IdP token endpoint, encoded as a JSON map.

redshift.auth.token_endpoint.request_url

Type: string.

The URL of the IdP token endpoint.

redshift.auth.token_type

Values: ACCESS_TOKEN, EXT_JWT.

The token type to use with IdP token authentication.

redshift.auto_create_user

Type: boolean. Default: false.

When connecting via IAM, whether to create the user if it does not exist.

redshift.aws.access_key_id

Type: string.

Explicitly supply the access key ID (by default the AWS SDK configures this, if needed).

redshift.aws.region

Type: string.

Explicitly supply the region (by default the AWS SDK configures this).

redshift.aws.secret_access_key

Type: string.

Explicitly supply the secret access key (by default the AWS SDK configures this, if needed).

redshift.aws.session_token

Type: string.

Explicitly supply the session token (by default the AWS SDK configures this, if needed).

redshift.aws.profile

Type: string.

Explicitly supply the profile to use (by default the AWS SDK configures this).

redshift.cluster_identifier

Type: string.

The cluster to connect to (only applicable if cluster_type is redshift or redshift-iam).

redshift.cluster_type

Values: redshift, redshift-iam, or redshift-serverless.

The type of Redshift cluster we are connecting to. redshift and redshift-iam both refer to Redshift clusters, while redshift-serverless means Redshift Serverless. They defer in how various other connection parameters are validated/inferred; see the connection documentation above.

redshift.connect_timeout_ms

Type: integer.

A connection timeout, in milliseconds.

redshift.custom_domain_name

Type: string.

Specify the workgroup to use by its custom domain name.

redshift.db_groups

Type: string.

A list of groups that the user should join on connection, as a JSON-serialized list of strings.

redshift.db_name

Type: string.

The name of the database to connect to.

redshift.ssl_cert, redshift.ssl_key, redshift.ssl_mode, redshift.ssl_root_key

Configure SSL/TLS options. See the Redshift documentation for details.

redshift.workgroup_name

Type: string.

Specify the workgroup to use by its name.

Bulk Ingest Options¶

redshift.ingest.bucket

Type: string

The S3 bucket to use as a staging area when ingesting data.

redshift.ingest.geo_type

Values: GEOMETRY, GEOGRAPHY. Default: GEOMETRY.

Whether GeoArrow columns (geoarrow.wkb, geoarrow.wkt) are ingested as GEOMETRY (default) or GEOGRAPHY.

redshift.ingest.geoarrow_wkb_size

Type: integer. Default: 1048576

The size of VARBYTE columns created during ingest, used as staging columns for GeoArrow WKB data.

redshift.ingest.varbyte_size

Type: integer. Default: 65536

The size of VARBYTE columns created during ingest.

Caveats¶

Temporary tables are shown in GetObjects, however, the order of the columns of these tables is not guaranteed to be correct. Also, less column/table metadata is returned.

Previous Versions¶

To see documentation for previous versions of this driver, see the following: