perf: Speedup to params simulation

[ValarDragon]

Description

In Osmosis' simulation, we were observing that a large percent of the time was spent within SimulateParamChangeProposalContent, in particular within the composed key function. This is from a trace of our simulator running for 10 minutes:

This change lowers the number of calls to ComposedKey() by a factor of 3, and speeds up the underlying ComposedKey function. (sprintf is not particularly efficient for string concatenation, https://dev.to/pmalhaire/concatenate-strings-in-golang-a-quick-benchmark-4ahh ). After these change, the effect of ComposedKey() can no longer be seen in the profiler outputs.

This change does not change the result of the simulator on a given seed.

See #9481 (comment) for description of changes.

---

Author Checklist

All items are required. Please add a note to the item if the item is not applicable and
please add links to any relevant follow up issues.

I have...

• included the correct type prefix in the PR title
• added ! to the type prefix if API or client breaking change
• targeted the correct branch (see PR Targeting)
• provided a link to the relevant issue or specification
• followed the guidelines for building modules
• included the necessary unit and integration tests
• added a changelog entry to CHANGELOG.md
• included comments for documenting Go code
• updated the relevant documentation or specification
• reviewed "Files changed" and left comments if necessary
• confirmed all CI checks have passed

Reviewers Checklist

All items are required. Please add a note if the item is not applicable and please add
your handle next to the items reviewed if you only reviewed selected items.

I have...

• confirmed the correct type prefix in the PR title
• confirmed ! in the type prefix if API or client breaking change
• confirmed all author checklist items have been addressed
• reviewed state machine logic
• reviewed API design and naming
• reviewed documentation is accurate
• reviewed tests and test coverage
• manually tested (if applicable)

[ValarDragon]

In a future PR, I suggest also making the simulator not try to bundle so many parameter change requests together.

I don't see how the test-cosmovisor failure is related to this change, as this code does not touch the state machine.

[alexanderbez]

Great find 👌

[ValarDragon]

Oh wait, this introduces a bug, I misunderstood the inner loop. (I thought if it was not ok, it just got skipped)

I'm just going to change this overall to remove the need for the inner loop, which originally made this function run in time O(num_params log(num_params)). It now runs in time O(num_params), and each operation is significantly quicker.

[ValarDragon]

So in the last two commits the following were changed:
Commit 1:
First title and desc were made to be more human readable, and smaller, in order to have better debuggability and lower entropy consumption. (This was a significant overhead)

Commit 2:

1. Fix the infinite loop bug introduced in commit 1.
2. Change the number of iterations from being a coupon collector problem to now being number of parameter changes.
3. Bound the number of parameter changes in one proposal to at most be 1000 changes

[ValarDragon]

Let me re benchmark whether or not this overall helps. (I believe each constituent change should help. The only dubious change is Change the number of iterations from being a coupon collector problem to now being number of parameter changes., but doing that removed the need for the expensive string concatenation / allocations from this loop which had (relatively) high overheads.

[ValarDragon]

This dramatically sped up our usecase. The same simulation run as before which timed out at 10 minutes on block 10, now finishes executing block 20 in under 10 seconds.

[codecov]

Codecov Report

Merging #9481 (a85b55d) into master (6e24a30) will increase coverage by 0.00%.
The diff coverage is 84.21%.

❗ Current head a85b55d differs from pull request most recent head 0caac5a. Consider uploading reports for the commit 0caac5a to get more accurate results

@@           Coverage Diff           @@
##           master    #9481   +/-   ##
=======================================
  Coverage   60.58%   60.59%           
=======================================
  Files         589      589           
  Lines       37218    37222    +4     
=======================================
+ Hits        22548    22553    +5     
+ Misses      12726    12725    -1     
  Partials     1944     1944           

Impacted Files	Coverage Δ
x/params/simulation/operations.go	86.36% <83.33%> (+8.58%)	⬆️
x/simulation/params.go	26.08% <100.00%> (ø)

[tac0turtle]

Noice

[tac0turtle]

@ValarDragon could you resolve conflicts, then the bot will take care of the merge