Measure WMS performance
Since performance and scalability is critical for agri, we should deploy a first prototype soon and measure the performance of different steps (tracing requests through the services, plus tracking run time of the evaluation steps in terravis as log output as well as metrics).
The goal is to be able to handle 2400 requests per minute with a good average and reasonable p99 performance. Note that in production the requests are not evenly distributed.
We should do this soon, because if it turns out that it's too slow for agri, we might have change the architecture or not use stacture for agri.