At the moment I think this will time the execution of the entire test, but I think we want to change that to be able to ignore setting up (e.g. by placing some random initial data in a bucket).
The way we use the run calling function that applies the benchmark context managers means that I think we are now safely not timing the setup or teardown, only the computing of the result.
At the moment I think this will time the execution of the entire test, but I think we want to change that to be able to ignore setting up (e.g. by placing some random initial data in a bucket).
_Originally posted by @TomNicholas in https://github.com/cubed-dev/cubed-benchmarks/pull/1#discussion_r1508275340_