Creating a DATALAKE Object - Teradata Vantage

The DATALAKE object encapsulates all the information needed to connect to an OTF data lake including the Authorization information needed to connect to the Catalog & Object storage and the connection details.

A DATALAKE object is created using the CREATE DATALAKE statement. All DATALAKEs are created in the TD_SERVER_DB database.

The Authorization information to connect to the Catalog and Object Storage is specified in the <auth_list> clause of the CREATE DATALAKE statement. See the topic Creating Authorization Objects for a DATALAKE for a DATALAKE on how to create AUTHORIZATION objects for a DATALAKE.

CREATE DATALAKE <datalake_name>
          <auth_list>
          USING catalog_type <left_paren> <catalog_type_literal> <right_paren>
                [catalog_location <left_paren> <catalog_location_literal> <right_paren>] 
                [storage_location <left_paren> <storage_location_literal> <right_paren>]
                [storage_region <left_paren> <storage_region_literal> <right_paren>]     
                [unity_catalog_name <left_paren> <unity_catalog_name_literal> <right_paren>]
                [storage_account_name <left_paren> <storage_account_name_literal> <right_paren>]
                [tenant_id <left_paren> <tenant_id_literal> <right_paren>]
                [default_cluster_id <left_paren> <default_cluster_id_literal> <right_paren>]
                [<custom_clause_list>]
           TABLE FORMAT <format_type>;
<datalake_name>::= !! Teradata identifier
<auth_list>::= EXTERNAL SECURITY <auth_type> [{<comma> <auth_list>}...]
<auth_type>::= DEFINER TRUSTED <connection_type> <auth_name> | [INVOKER] TRUSTED <connection_type> <auth_name>
<connection_type>::= CATALOG | STORAGE
<auth_name>::= !! Teradata identifier
<catalog_type_literal>::= <quote> <catalog_type> <quote>
<catalog_location_literal>::= <quote> <catalog_uri> <quote>
<storage_location_literal>::= <quote> <storage_uri> <quote>
<storage_region_literal>::=  <quote> <storage_region> <quote>
<unity_catalog_name_literal>::=  <quote> <unity_catalog_name> <quote>
<storage_account_name_literal>::=  <quote> <storage_account_name> <quote>
<tenant_id_literal>::=  <quote> <tenant_id> <quote>
<catalog_type>::= hive | glue | unity
<catalog_uri>::= !! Catalog URI, e.g. thrift://example.com, required when catalog_type is hive or unity
<storage_uri>::= !! Storage URI, e.g. s3://example-iceberg-v1/, required when catalog_type is hive
<storage_region>::= !! Cloud region, e.g. us-west-2, required when catalog_type is glue or hive
<unity_catalog_name>::= !! Unity/Azure Databricks catalog name, e.g. reg_iceberg_db, required when catalog_type is unity
<storage_account_name>::= !! Azure DataLake Storage Gen2 storage account name, e.g. regicebergstorageacct, required when catalog_type is unity
<tenant_id>::= !! Azure Active Directory service principal tenant ID, e.g. 391c8c4c-6a2a-40fd-ab98-226b6baa5155, required when catalog_type is unity
<default_cluster_id>::= !! Spark compute cluster ID; the expected format is xxxx-xxxxxx-xxxxxxxx, e.g. 0210-232334-ab0q59t3, required when catalog_type is unity
<custom_clause_list>::= <custom_clause> [<custom_clause_list>...]
<custom_clause>::= <name> <left_paren> <quote> <value>[{<comma><value>}...] <quote> <right_paren>
<name>::= !! Teradata identifier
<value>::= !! Teradata literal value
<format_type>::= iceberg | deltalake

Usage Considerations

Catalog Location is missing or
                invalid.

bfss://otf-330spark-51hdi-publ-2024-06-13t06-52-39-186z@iceberg-storageeastus2-.dfs.core.-windows.net/s3://vim-iceberg-v1/

The procedure to create DATALAKE objects for different catalogs & object storages is discussed in the following topics.