Andrew Fitzgerald

Happy to add additional bench settings for testing the different implementations in a consistent way. Also happy to add some variability in the transaction execution time for these simple cases

code to benchmark different scheduling algorithms/implementations for #27027

This was an initial idea for: #27027

Testing on MNB snaphot(s): - Full snapshot: snapshot-141271324-5S45gzue6kS5ffkeFrF23CX1tbdEGi8yUmucfeuhS52V.tar.zst - Incremental snapshot: incremental-snapshot-141271324-141290659-9KTPeHb81ZjnAYTbqn9bmv9kAyNJR6SESwMWbqxXrH4z.tar.zst | | Full Snapshot Untar (s) | Incremental Snapshot Untar (s) | Rebuild Bank (s) | Total (s)...

Tested with multiple account directories, worked as expected.

This creates a potential bottleneck since we're moving from N banking threads doing deserialization to a single thread for deserializing. Need to benchmark this change to make sure this will...

@taozhu-chicago sorry had to rebase because of conflict in adding features to feature-set

> > Just a couple of minor comments, feel free to just do them as follow-up PRs. Appreciate all the back-and-forth on the PR! > > thanks for timely review...

I like this approach, but we should be a bit careful with the locks on shared resources: cost_tracker, account_locks 1. we weant to minimize the time we hold these locks...

> Yeah taking both locks is probably too risky.. How about we first reserve as much as cost as we need in qos and then keep track of how much...