Tuning gas schedule based on actual throughput performance

[igor-aptos]

Based on the single-threaded benchmarking of all workloads in the single_node_performance.py, reduce the variance from 10x to under 1.5x

Description

Test Plan

[igor-aptos]

[!WARNING] This pull request is not mergeable via GitHub because a downstack PR is open. Once all requirements are satisfied, merge this PR as a stack on Graphite. Learn more

Current dependencies on/for this PR:

• main
  • PR #11292
    • PR #11237
      • PR #11254
        • PR #11017
          • PR #11130 👈
      • PR #11114

This stack of pull requests is managed by Graphite.

[igor-aptos]

will share the results from #11017, but thus far looks very promising that these are the only changes needed

[msmouse]

I'm all for it, but every txn will be more expansive after this change (gas for no-op txn bumps from 2 to 6), how do we green light this?

[igor-aptos]

Will bring up for a discussion but:

• this increases only cheap transactions (coin transfer from 6 to 10), but reduces price for both reads and writes.
• we only care about absolute values for calibration, so we could just take all values and divide by 3. (as well as block/transaction gas limits)

[github-actions[bot]]

This issue is stale because it has been open 45 days with no activity. Remove the stale label, comment or push a commit - otherwise this will be closed in 15 days.