This paper benchmarks ingestion from PostgreSQL, MySQL, SQL Server, and Kafka, from initial snapshot through steady-state throughput to vertical and horizontal scaling.

The results show that Materialize sustains high freshness and throughput with predictable load on upstream systems, while maintaining strong consistency guarantees.

While these results are a useful starting point, the best way to assess Materialize is to run a proof of concept on your own data and hardware. Reach out to the Materialize team, and we can help you find the optimal setup for your requirements and workload.