Pentaho Consulting, Support & Migration Services

It depends on your batch volume, how many undocumented flows you're carrying, and your exposure to Hitachi Vantara's licensing and support changes. We give you a written assessment based on your actual environment — for teams with heavy, well-understood batch logic and no urgent license pressure, stabilizing in place can be the right near-term call. For most others, migration to Airbyte/dbt or a lakehouse pays back within a year.

Hitachi Vantara has shifted licensing terms and support scope for Pentaho over recent years, and some customers have found their support path unclear or their renewal costs rising without a corresponding increase in platform investment. We help you assess your specific contract and support-gap exposure, then remediate or plan around it.

Yes — this is one of our most common engagements. We reverse-engineer transformation and job dependencies directly from the .ktr/.kjb files and repository metadata, producing source-to-target lineage documentation your team can use for compliance, onboarding, or a future migration decision.

We map each transformation's logic to an equivalent extraction step in Airbyte and a transformation model in dbt, run both pipelines in parallel to validate output parity, then cut over in phases so no nightly job goes dark. Complex custom Java steps in Kettle sometimes require the most careful translation — we flag those early.

Yes. We stabilize and tune Pentaho Server and Carte slave clusters for distributed batch execution, including load distribution, memory tuning per slave, and dependency documentation for jobs that assume a specific execution order.

We start by isolating whether the failure is a data issue (bad source row, schema drift) or an infrastructure issue (memory, disk, database contention), since Pentaho's error messages often point at the wrong layer. We stabilize the job, backfill any missed run, and document the root cause so it doesn't recur.