May 2026

Run edge processing entirely on infrastructure you own and operate. You manage the deployment lifecycle end-to-end; Ataccama provides the software, Terraform artifacts, and upgrade packages.

Self-managed deployment is the right fit when you need:

The Ataccama-managed AWS deployment remains available for teams who prefer Ataccama to provision and operate the edge compute for them.

For deployment instructions, see Deploy a Self-Managed Edge. For an overview of both options, see Edge Processing.

Pull database passwords, API keys, and other credentials directly from AWS Secrets Manager when configuring edge data sources, keeping credential management in your existing AWS infrastructure. Supports both same-account and cross-account setups.

Set up an IAM role that grants the edge read access to your secrets, then add a key vault connection in ONE pointing at that role. From there, reference secrets by name when configuring source credentials.

For details, see Set Up AWS Secrets Manager Access for Edge.

Bring Your Own Key (BYOK) lets you encrypt your Ataccama ONE environment’s data at rest using AWS KMS keys you create and control in your own AWS account. Key material never leaves your account, and every KMS operation is logged in your own CloudTrail for full auditability.

BYOK protects both data stores that hold your environment data:

Each environment gets its own dedicated pair of keys, providing cryptographic isolation between environments.

Configure BYOK in the Create SaaS environment wizard in Ataccama Cloud Portal — a CloudFormation template sets up both keys for you.

For details, see Bring Your Own Key (BYOK).

You can now generate an AI assessment directly on any alert to quickly understand what happened, what’s affected, and what it means for your data.

For details, see Investigate Alerts.

You can now use the AI Agent to investigate and manage alerts.

For details, see AI Agent.

Profile data and evaluate DQ rules directly on your Databricks cluster — your data never leaves the source.

Instead of moving data into Ataccama ONE, the platform generates SQL that runs inside Databricks using its compute. Only results (and invalid samples, if configured) come back.

What you get:

What to know before you turn it on:

For setup details, see Databricks Connection.

Configure each data source once and reuse it everywhere. Scan plans now inherit connection details from the same connection you set up for data processing, instead of requiring connection parameters declared separately in the scan plan.

The result:

For details, see Built-In Lineage Scanner.

You can now grant reference data table roles to user groups, not just individual users. When a group holds a role, every current and future member inherits the role’s permissions — useful for teams that need the same level of access on a table.

The Request review form now also includes a Reviewer field that lists users and groups with the Approver or Owner role on the table. The first eligible candidate is preselected — change it to route the review to a specific person or team.

For details, see Assign roles to groups and Publish changes.

You can now view data quality results and run evaluations on Published and In review reference data datasets, not just Draft.

For In review datasets, the results reflect the future state of the published dataset — the data quality it will have once the in-review changes are applied.

For details, see Monitor Data Quality of Reference Data.

Pick values directly from the full source table when editing connected attributes, with filtering to navigate large datasets. Now available in the record detail and bulk edit dialogs, in addition to the data grid.

For details, see Select from the source table.

Filter the records in a reference data table by their data quality validation status to focus on the records that need attention.

When you filter a table, you can now show only:

The filter uses the results of the most recent data quality evaluation.

For details, see Filter records.