Bring Real-Time Data to Your Data Lakehouse and AI

Stream data from legacy and modern systems, making fresh, governed data available to your lakehouse without building separate pipelines for every use case.

Streamhouse + Lakehouse: Better Together

Your lakehouse is built for analytical workloads. Your Streamhouse keeps changing business data current, governed, and ready for real-time action.

Together, they connect operational and analytical data—so AI agents, applications, and analytics can work with trusted business context without requiring every workload to pass through a centralized analytical store.

Eliminate data silos. Do away with outdated, inefficient workflows that silo data from low throughput and high throughput sources, eliminating costly and difficult-to-manage replication of data.

Free your data. Ensure interoperability across clouds, warehouses, query engines and open-source data processing frameworks.

Scale performance, not cost. Cut your data infrastructure costs by 50% or more compared to traditional data platforms by running workloads incrementally on highly optimized compute.

Build with Confluent

This use case leverages the following building blocks in Confluent Cloud:

Reference Architecture




In this example, we are replicating a PostgreSQL database on AWS to the Onehouse Universal Data lake with a Debezium Connector and Confluent Kafka with up-to-the-minute data freshness.
Connect to PostgreSQL with Debezium

Debezium is an open-source distributed platform for change data capture (CDC).

Create a new schema in Confluent

Onehouse automatically creates a new schema with Confluent’s Schema Registry (Stream Governance). Onehouse manages the Debezium connector, Confluent, and the schema to optimize cost and reduce engineering overhead.

Run your continuous pipeline

Onehouse will ping PostgreSQL for any inserts, updates, and deletes at the interval you’ve set for up-to-the minute data. You can perform transformations such as masking and deduplication as part of your pipelines, and Onehouse enables you to create validations and quarantine suspicious data, so you can ensure data quality at the point of ingestion.

Query your replicated data

Onehouse integrates with a number of data catalogs so you can query and process your data downstream with the best tool for the job.

Resources

Book an Expert Consult

"제출"을 클릭하면 Confluent에서 보내는 마케팅 메일 수신에 동의하게 됩니다. 또한 사용자의 개인정보가 당사의 개인정보 보호정책에 따라 처리되는 것에 동의합니다.