Batch throughput with concurrent batch requests

[aequitas]

When multiple batch requests are started simultaneously it seems only the first (few) requests run at full speed. The rest of the requests run at a reduced speed, only using part of the resources available.

It might also have to do with caching, since these runs were all with the same dataset, so some results might be gotten from cache instead of calculated giving lower resource usage. Investigation is needed to determine if the total runtime of the batch requests it actually longer.

[gthess]

From what I remember caching plays a big part in batch. I believe if the different batch requests have unique domains to be tested (this could be difficult to do with nameserver and mailserver setups that are using the same third party services, which could have cached results) you may see a different picture.