Data Lakehouse
Query your lakehouse without copying the data.
Apache Iceberg made the lakehouse possible. PhoenixAI makes it fast. Query your lakehouse tables at sub-second latency — no ETL, no data movement, no separate warehouse to maintain.
The problem
The lakehouse is great for storage. Not always fast for queries.
The lakehouse query problem
Slow interactive queries force data movement
- Complex analytical queries on Apache Iceberg take minutes, not seconds
- Teams copy data into a warehouse just to get interactive performance
- Two copies of data: one in the lake, one in the warehouse
- Sync pipelines between lake and warehouse break and drift
With PhoenixAI
Data warehouse speed. Data lake economics.
- Sub-second queries directly on Apache Iceberg tables
- Intelligent tiered cache — no copy needed, warehouse-like speed
- Single source of truth — no sync pipeline, no data drift
- Query streaming and historical in the same query
Capabilities
The query layer your lakehouse needs.
What PhoenixAI adds to an Apache Iceberg architecture.
Native lakehouse integration
Native execution on lakehouse tables such as Apache Iceberg delivers data warehouse performance, directly on your lake tables — no copies required.
Intelligent tiered cache
Tiered cache across memory and local SSD delivers sub-second Apache Iceberg queries, with all your data persisted in the lake as a single source of truth.
Query streaming and historical in one query
Join or union real-time streaming data with historical lakehouse tables in the same SQL query. No federation overhead, no separate engine for each layer.
Async materialized views
Pre-compute hot data on Apache Iceberg tables. Queries automatically rewrite to hit the MV instead of scanning the full Iceberg table — sub-second answers on multi-petabyte fact tables.
Works with your lakehouse
PhoenixAI integrates with open table formats, object storage, catalogs, and the processing engines that feed your lakehouse.
Open table formats
- Apache Iceberg
- Delta Lake
- Apache Hudi
Object storage
- Amazon S3
- Google GCS
- Azure ADLS
- MinIO
Catalogs & platforms
- AWS Glue
- Hive Metastore
- Databricks Unity Catalog
- Snowflake Horizon Catalog
- Amazon S3 Tables
In production
What lakehouse teams say after switching.
200ms avg. query response
We provide booking services for over 1.5 million hotels worldwide. By using PhoenixAI we realized high-speed data analysis with an average query response speed of 200ms. Thanks to the unified data analytical architecture, manpower and hardware costs are greatly reduced.
Query your lakehouse in real time.
Connect PhoenixAI to your Apache Iceberg tables. We'll show you what your query latency looks like.