Is your feature request related to a problem? Please describe.
The data formats, JSON & XML, only profile the entire dataset to test how this library handles large data sets.
This is something we'd like to keep, however, perhaps it would also make sense to benchmark independent DataNodes with normal benchmarking. That way it's possible to monitor subtle performance impact.
What may be important here, is that the larger benchmark has a custom runner made for that purpose.
Perhaps it's smart to make a second one that's purpose built for normal benchmarking?
And maybe we need separate benchmarking csproj or a way to select what to benchmark.
Is your feature request related to a problem? Please describe. The data formats,
JSON
&XML
, only profile the entire dataset to test how this library handles large data sets. This is something we'd like to keep, however, perhaps it would also make sense to benchmark independent DataNodes with normal benchmarking. That way it's possible to monitor subtle performance impact.What may be important here, is that the larger benchmark has a custom runner made for that purpose. Perhaps it's smart to make a second one that's purpose built for normal benchmarking? And maybe we need separate benchmarking
csproj
or a way to select what to benchmark.