- Home
- Data Lakehouse Platforms
- Google Cloud Lakehouse for Apache Iceberg
Google Cloud Lakehouse for Apache Iceberg
Mountain View, California, United States
What it is
Google Cloud Lakehouse for Apache Iceberg is a web offering for working with Apache Iceberg tables. Its listed capabilities include SQL analytics, ACID transactions, streaming ingestion, and a governance catalog. It supports both listed table-format categories and both storage models. Pricing is divided among metadata storage, table management, and metadata access. A metadata storage free tier includes 1 GiB per month; the listed storage plan starts at $0.04 per GiB per month and also includes 1 GiB free. Table management starts at $0.12 per DCU-hour and includes automatic table storage optimization, adaptive file sizing, automatic clustering, garbage collection, and BigQuery metadata generation. Class A metadata access includes 5,000 free operations monthly, and Class B includes 50,000. Catalog federation metadata refresh lists 0 to 5,001 free operations per account each month. The service is paid overall and has no free plan. Google was founded in 1998 and is headquartered in Mountain View, California.
For and against
For
- Includes SQL analytics and ACID transactions.
- Supports streaming ingestion.
- Includes a governance catalog.
- Metadata storage has a 1 GiB monthly free tier.
- Table management includes automatic clustering and garbage collection.
Against
- No free plan is listed.
- Table management starts at $0.12 per DCU-hour.
- Metadata storage beyond the included amount starts at $0.04 per GiB per month.
The verdict
Google Cloud Lakehouse for Apache Iceberg combines table management and analytics capabilities with separately described metadata charges. It has a metadata storage free tier, but the overall service is paid and has no free plan.
Who it suits
This offering may suit organizations working with Apache Iceberg tables that need SQL analytics, streaming ingestion, ACID transactions, and a governance catalog.
Plans and prices
Metadata storage free tier
1 GiB metadata storage
Read 20 Sep 2026 · cloud.google.com
Metadata storage
1 GiB free metadata storage included
Read 20 Sep 2026 · cloud.google.com
Table management
Automatic table storage optimization · Adaptive file sizing · Automatic clustering · Garbage collection
Read 20 Sep 2026 · cloud.google.com
Catalog federation metadata refresh
0 to 5,001 operations free per month per account
Read 20 Sep 2026 · cloud.google.com
Metadata storage included tier
Lakehouse runtime catalog metadata
Read 15 Oct 2026 · cloud.google.com
Metadata storage overage
metadata beyond included 1 GiB
Read 15 Oct 2026 · cloud.google.com
Lakehouse table management
automatic file sizing · clustering · garbage collection · BigQuery metadata generation
Read 15 Oct 2026 · cloud.google.com
Class B metadata access
read · delete · getIamPolicy
Read 15 Oct 2026 · cloud.google.com
Class A metadata access
create · register · list · update
Read 15 Oct 2026 · cloud.google.com
On the record
- AI features
- BigQuery ObjectRefs combine structured Iceberg data with unstructured objects for multimodal, vector and graph analytics.cloud.google.com · read 15 Oct 2026
- API
- The Lakehouse API is a serverless, highly available metastore API exposing REST resources for catalogs and data-sharing objects.docs.cloud.google.com · read 15 Oct 2026
- Automated maintenance
- Lakehouse can automate file sizing, compaction, clustering, garbage collection, partitioning and history-based optimization.cloud.google.com · read 15 Oct 2026
- BigQuery analytics
- BigQuery supports vectorized execution, partition pruning, multi-statement transactions and SQL DDL/DML over Lakehouse Iceberg tables.cloud.google.com · read 15 Oct 2026
- Company founding and headquarters
- Google says it was officially born in 1998 and moved its headquarters to Mountain View, California; LinkedIn lists Mountain View as headquarters.about.google · read 15 Oct 2026
- Company size
- LinkedIn lists Google as a public software-development company with more than 10,001 employees.linkedin.com · read 15 Oct 2026
- Compliance
- Google Cloud lists ISO 27001, ISO 27017, ISO 27018, ISO 27701, SOC 1/2/3, PCI DSS, FedRAMP, GDPR and HIPAA offerings or alignments.cloud.google.com · read 15 Oct 2026
- Cross-cloud federation
- Preview federation can expose data cataloged in AWS Glue, Databricks Unity Catalog and Snowflake Horizon Catalog to BigQuery and Spark without moving the data.cloud.google.com · read 15 Oct 2026
- Deployment model
- Lakehouse runtime catalog is fully managed and removes the need to operate a self-hosted Hive Metastore.docs.cloud.google.com · read 15 Oct 2026
- Developer documentation
- Google provides introductory tutorials, quickstarts, reference material, code samples, engine-specific guides and troubleshooting documentation.docs.cloud.google.com · read 15 Oct 2026
- Governance
- Knowledge Catalog integration supports centralized governance policies, semantic search, lineage and quality checks across engines.docs.cloud.google.com · read 15 Oct 2026
- Infrastructure as code
- The Google Terraform provider provisions Lakehouse runtime catalog resources, including Iceberg REST catalog endpoints.docs.cloud.google.com · read 15 Oct 2026
- Interoperability
- The Iceberg REST catalog provides read and write interoperability among BigQuery, Managed Service for Apache Spark, Spark, Flink, Trino and other compatible engines.cloud.google.com · read 15 Oct 2026
- Legal entity
- Google Cloud's contracting-entity page identifies Google LLC for the United States and other uncovered locations.cloud.google.com · read 15 Oct 2026
- License
- The Apache Iceberg project is distributed under the Apache License 2.0; Google Cloud Lakehouse itself is a proprietary managed service.github.com · read 15 Oct 2026
- Limit: API access
- The BigQuery tabledata.list method cannot retrieve rows directly from runtime-catalog tables.docs.cloud.google.com · read 15 Oct 2026
- Limit: cost visibility
- BigQuery dry runs can report zero processed bytes for runtime-catalog queries even though the executed query incurs processing cost.docs.cloud.google.com · read 15 Oct 2026
- Limit: orchestration and scheduling
- TrustRadius reviewers report that orchestration and scheduled-query workflows can be awkward and require practice.trustradius.com · read 15 Oct 2026
