Additional BigLake metastore features

To customize your BigLake metastore configuration, you can use the following additional features:

  • Apache Spark Iceberg procedures
  • The filter option for unsupported tables
  • BigQuery connection overrides
  • Access control policies for BigLake metastore Iceberg tables

Use Iceberg Spark procedures

To use Iceberg Spark procedures, you must include Iceberg SQL extensions in your Spark configuration. For example, you can create a procedure to roll back to a previous state.

Use interactive Spark-SQL to roll back to a previous state

You can use an Iceberg Spark procedure to create, modify, and roll back a table to its previous state. For example:

  1. Create a Spark table:

    spark-sql \
   --jars https://storage-download.googleapis.com/maven-central/maven2/org/apache/iceberg/iceberg-spark-runtime-3.5_2.12/1.6.1/iceberg-spark-runtime-3.5_2.12-1.6.1.jar,gs://spark-lib/bigquery/iceberg-bigquery-catalog-1.6.1-1.0.1-beta.jar \
   --conf spark.sql.extensions=org.apache.iceberg.spark.extensions.IcebergSparkSessionExtensions \
   --conf spark.sql.catalog.CATALOG_NAME=org.apache.iceberg.spark.SparkCatalog \
   --conf spark.sql.catalog.CATALOG_NAME.catalog-impl=org.apache.iceberg.gcp.bigquery.BigQueryMetastoreCatalog \
   --conf spark.sql.catalog.CATALOG_NAME.gcp_project=PROJECT_ID \
   --conf spark.sql.catalog.CATALOG_NAME.warehouse=WAREHOUSE_DIRECTORY

    Replace the following:

    • CATALOG_NAME: the catalog name that references your Spark table.

    • PROJECT_ID: the ID of the Google Cloud project.

    • WAREHOUSE_DIRECTORY: the URI of the Cloud Storage folder where your data warehouse is stored.

    USE `CATALOG_NAME`;
CREATE NAMESPACE NAMESPACE_NAME;
USE NAMESPACE NAMESPACE_NAME;
CREATE TABLE NAMESPACE_NAME.TABLE_NAME (id int, data string) USING ICEBERG LOCATION 'WAREHOUSE_DIRECTORY';
INSERT INTO NAMESPACE_NAME.TABLE_NAME VALUES (1, "first row");
DESCRIBE EXTENDED TABLE_NAME;

    Replace the following:

    • NAMESPACE_NAME: the namespace name that references your Spark table.
    • TABLE_NAME: a table name that references your Spark table.

    The output contains details about the table configuration:

    ...
Table Properties [current-snapshot-id=1659239298328512231,format=iceberg/parquet,format-version=2,write.parquet.compression-codec=zstd]
...

  2. Alter the table again, and then roll it back to the previously created snapshot 1659239298328512231:

    ALTER TABLE TABLE_NAME ADD COLUMNS (newDoubleCol double);
INSERT INTO TABLE_NAME VALUES (2, "second row", 2.5);
SELECT * FROM TABLE_NAME;
CALL CATALOG_NAME.system.set_current_snapshot('NAMESPACE_NAME.TABLE_NAME', SNAPSHOT_ID);
SELECT * FROM TABLE_NAME;

    Replace the following:

    • SNAPSHOT_ID: the ID of the snapshot you are rolling back to.

    The output is similar to the following:

    1 first row
Time taken: 0.997 seconds, Fetched 1 row(s)

Filter unsupported tables from table listing functions

When you use Spark SQL with the BigLake metastore catalog, the SHOW TABLES command shows all the tables in the specified namespace, even those that aren't compatible with Spark.

To only display supported tables, turn on the filter_unsupported_tables option:

spark-sql
  --jars https://storage-download.googleapis.com/maven-central/maven2/org/apache/iceberg/iceberg-spark-runtime-3.5_2.12/1.6.1/iceberg-spark-runtime-3.5_2.12-1.6.1.jar,gs://spark-lib/bigquery/iceberg-bigquery-catalog-1.6.1-1.0.1-beta.jar \
  --conf spark.sql.catalog.CATALOG_NAME=org.apache.iceberg.spark.SparkCatalog \
  --conf spark.sql.catalog.CATALOG_NAME.catalog-impl=org.apache.iceberg.gcp.bigquery.BigQueryMetastoreCatalog \
  --conf spark.sql.catalog.CATALOG_NAME.gcp_project=PROJECT_ID \
  --conf spark.sql.catalog.CATALOG_NAME.gcp_location=LOCATION \
  --conf spark.sql.catalog.CATALOG_NAME.warehouse=WAREHOUSE_DIRECTORY \
  --conf spark.sql.catalog.CATALOG_NAME.filter_unsupported_tables="true"

Replace the following:

  • CATALOG_NAME: the name of the Spark catalog to use.
  • PROJECT_ID: the ID of the Google Cloud project to use.
  • LOCATION: the location of the BigQuery resources.
  • WAREHOUSE_DIRECTORY: the Cloud Storage folder to use as the data warehouse.

Set a BigQuery connection override

You can use BigQuery connections to access data stored outside of BigQuery, such as in Cloud Storage.

To set a BigQuery connection override that provides access to a Cloud Storage bucket, complete the following steps:

  1. In your BigQuery project, create a new connection to your Cloud Storage resource. This connection defines how BigQuery accesses your data.

  2. Grant the user or service account accessing the data the roles/bigquery.connectionUser role on the connection.

    Make sure that the connection resource shares the same location as the target resources in BigQuery. For more information, see Manage connections.

  3. Specify the connection in your Iceberg table with the bq_connection property:

    CREATE TABLE TABLE_NAME (id int, data string) USING ICEBERG LOCATION 'WAREHOUSE_DIRECTORY' TBLPROPERTIES ('bq_connection'='projects/PROJECT_ID/locations/LOCATION/connections/CONNECTION_ID');

    Replace the following:

    • TABLE_NAME: a table name for your Spark table.
    • WAREHOUSE_DIRECTORY: the URI of the Cloud Storage bucket that stores your data.
    • PROJECT_ID: the ID of the Google Cloud project to use.
    • LOCATION: the location of the connection.
    • CONNECTION_ID: the ID of the connection.

Set access control policies

You can enable fine-grained access control (FGAC) on BigLake metastore Iceberg tables by configuring access control policies. You can only set access control policies on tables that use a BigQuery connection override. You can set these policies in the following ways:

After you configure your FGAC policies, you can query the table from Spark using the following example:

from pyspark.sql import SparkSession

# Create a Spark session
spark = SparkSession.builder \
.appName("BigLake Metastore Iceberg") \
.config("spark.sql.catalog.CATALOG_NAME", "org.apache.iceberg.spark.SparkCatalog") \
.config("spark.sql.catalog.CATALOG_NAME.catalog-impl", "org.apache.iceberg.gcp.bigquery.BigQueryMetastoreCatalog") \
.config("spark.sql.catalog.CATALOG_NAME.gcp_project", "PROJECT_ID") \
.config("spark.sql.catalog.CATALOG_NAME.gcp_location", "LOCATION") \
.config("spark.sql.catalog.CATALOG_NAME.warehouse", "WAREHOUSE_DIRECTORY") \
.getOrCreate()

spark.sql("USE `CATALOG_NAME`;")

# Configure spark for storing temp results
spark.conf.set("viewsEnabled","true")
spark.sql("CREATE namespace if not exists MATERIALIZATION_NAMESPACE");
spark.conf.set("materializationDataset","MATERIALIZATION_NAMESPACE")

spark.sql("USE NAMESPACE DATASET_NAME;")

sql = """SELECT * FROM DATASET_NAME.ICEBERG_TABLE_NAME"""
df = spark.read.format("bigquery").load(sql)
df.show()

Replace the following:

  • CATALOG_NAME: the name of your catalog.
  • PROJECT_ID: the ID of the project that contains your BigQuery resources.
  • LOCATION: the location of the BigQuery resources.
  • WAREHOUSE_DIRECTORY: the URI of the Cloud Storage folder that contains your data warehouse.
  • MATERIALIZATION_NAMESPACE: the namespace where you want to store temporary results.
  • DATASET_NAME: the name of your dataset that contains the table that you are querying.
  • ICEBERG_TABLE_NAME: the name the table that you are querying.

What's next