On August 17, 2026, Inova announced an update to Aurora, its AI infrastructure for public health. The stated scope covers ingestion, cataloging, and preparing datasets before analysis; the announcement provides no performance metrics or general commercial availability details.
Data Hub catalogs JSONL files, tracks processing, and supports queries on structure and records. Uploads can be sent in parts; metadata such as size, row count, and checksum is recorded. Exploration supports search, cursors, sorting, and filters, with a documented limit of 100 records per query.
After validation, datasets enter an ingestion queue. Content goes to the ClickHouse analytical warehouse, while the catalog and metadata remain in MySQL. States such as “queued,” “ingesting,” “ready,” and “ingest_failed” show progress; “ready” means the dataset is available for exploration in the warehouse.
Data Quality works on ready datasets: users can select filters and columns, create routines, and run analyses. The first described routine applies a regular expression to a field and tags matching records. To assess the announcement, consult the original publication and module documentation; verify limits, states, and permissions, and do not assume availability beyond what was stated.