SecurityBrief Canada - Technology news for CISOs & cybersecurity decision-makers
Canada
Rubrik launches Iceberg protection for AWS lakehouses

Rubrik launches Iceberg protection for AWS lakehouses

Tue, 8th Sep 2026 (Today)
Joseph Gabriel Lagonsin
JOSEPH GABRIEL LAGONSIN News Editor

Rubrik has launched Apache Iceberg Protection for Apache Iceberg on AWS, aimed at organisations that need to recover Iceberg tables after a cyber incident.

The offering restores complete, queryable tables rather than only the underlying files. It backs up both table data and the metadata that defines the tables, then reconnects the catalogue during recovery.

That addresses a problem in lakehouse environments, where native snapshots act as metadata pointers rather than full backups. If a table is deleted, corrupted or overwritten, those pointers can disappear with the data, leaving teams to restore files first and then manually rebuild metadata before the data can be queried again.

The product covers Iceberg tables held in AWS Glue Data Catalog and Amazon S3 Tables. Recovered tables can be queried immediately in Amazon Athena, Apache Spark and Trino.

Lakehouse focus

The launch extends Rubrik's data security business further into lakehouse environments, which have become more central to analytics, AI workloads and broader business operations. Apache Iceberg, an open table format for large analytical datasets, has gained traction among companies building data platforms on cloud object storage.

The new offering also reflects a shift in backup requirements for modern data estates. Protecting raw object storage alone may not be enough when production systems depend on catalogues, metadata layers and table definitions to make data usable.

Customers can store immutable backups within their own AWS accounts, which Rubrik positions as a data sovereignty option. It also offers storage through Rubrik Cloud Vault, its managed air-gapped backup service.

The service uses forever-incremental backups of the latest compacted snapshots to keep backup windows manageable for very large deployments. Customers can also choose the Amazon S3 storage class used for backups.

Recovery process

Beyond cyber recovery, the product can also be used before operational changes that carry risk, such as schema modifications or large batch writes. In those cases, users can create an on-demand backup and restore it to an isolated Iceberg branch to check data and schema before moving the table into production.

That approach targets a wider range of recovery scenarios than ransomware alone. Data teams often need point-in-time restoration after accidental deletion, faulty automation or changes introduced by internal tools and AI systems.

Anneka Gupta, Chief Product Officer at Rubrik, outlined the company's position on the launch.

"Data lakes have grown beyond analytics and AI to now run the entire business, and Apache Iceberg is one of the most critical systems built on them," Gupta said. "While Rubrik has long protected the underlying object storage, we now protect the lakehouse itself. That means recovering complete, queryable tables, not just raw files, with immutable backups stored directly in your own AWS account. By unifying these capabilities on a single platform, the same policies protect the lakehouse alongside the surrounding databases, pipelines, and code."

The product sits within Rubrik's broader cloud data security and AI governance portfolio on AWS. The company has been expanding from backup and recovery into a broader set of tools focused on cyber resilience, data control and operational recovery across cloud environments.

For enterprises running Apache Iceberg at scale, the issue is less about retrieving data files than restoring a working table state quickly enough to avoid prolonged disruption. In practice, that means recovery must include metadata, catalogue registration and compatibility with query engines already used by analysts and applications.

Rubrik's pitch is that Iceberg-aware recovery can reduce the manual work involved after an incident. The significance of the launch will depend on how widely organisations have standardised on Apache Iceberg for production datasets, and whether they prefer to keep backup copies inside their own AWS accounts rather than in vendor-managed infrastructure.