aggregateUTxOCoinByCredential benchmarks and performance tuning #2241
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
It was only after writing this that I noticed we already had some benchmarks for this function, in - https://github.com/input-output-hk/cardano-ledger-specs/blob/master/shelley/chain-and-ledger/shelley-spec-ledger-test/bench/BenchUTxOAggregate.hs and - https://github.com/input-output-hk/cardano-ledger-specs/blob/master/shelley/chain-and-ledger/shelley-spec-ledger-test/bench/Main.hs#L221 However, upon examining that benchmark, I think it's still worth having this one (and maybe dropping the other). The existing benchmark uses `gen` (I believe without fixing the seed), making it hard to test the same case each time. And whilst we can control the size of the UTxO, this new benchmark allows explicit control over the key things important to the UTxO aggregation calculation, namely: - The number of base stake addresses - The number of pointer stake addresses (which require some additional computation), and - The "delegation" factor; i.e. the number of TxOuts for each stake address.
This makes a couple of minor improvements in the function 'aggregateUtxoCoinByCredential', as measured by the new benchmark introduced. I tried a few variants; see https://drive.google.com/drive/u/1/folders/1o7r4EF0uM5gfhtf3_GoMSJdoHc_d6Xtc for the full reports. This version seems to shave maybe 20% off of the 20K UTxO run. Interestingly, a variant where we try to explicitly deserialise only the stake ref (using `deserialiseAddrStakeRef`) was _considerably_ slower. I'm not immediately sure why!
|
Some subsidiary results from this:
|
|
But also: I haven't managed to actually be patient enough to get the benchmark results for a UTxO with 200,000 stake addresses yet - this definitely gets very expensive quickly. |
JaredCorduan
approved these changes
Apr 19, 2021
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
I took a look at this a bit on Friday morning, and given the brouhaha over the weekend, it seemed worth pushing!
This PR adds a more detailed and comprehensive benchmark for
aggregateUTxOCoinByCredential, and makes some (minor) performance improvements.