The code is written in high performance multi-threaded Rust and uses the [1] Arrow compute framework. We also batch events and target about 32MB of event data per lambda invocations. As a result it can process tens of thousands of events per second per thread, depending on the number of transformations.
That said, we are working on performance estimates and a benchmark on some real world data for Matano to help users like you better understand the cost factors. Stay tuned.