Conversation
…ERUN_CMD) Runs an operator command between 'pre-run steps complete, client stopped, fs synced' and 'schelk promote', so baseline surgery gets baked into the promoted virgin. Motivating use: draining geth's pathdb journal into the disk layer (drainjournal tool) so no benchmark test inherits journal-resident, RAM-served trie state - the root cause of the DIFF_MAX divergence (state-db root-cause ledger R1-R3). Hook failure aborts before promote: a half-treated baseline must never replace the golden image.
This file contains hidden or 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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Runs an operator command between "pre-run steps complete, client stopped, fs synced" and
schelk promote, so baseline surgery gets baked into the promoted virgin. Hook failure aborts before the promote, so a half-treated baseline can never replace the golden image.Motivation. A baseline built by replaying a pre-run bundle keeps the last ~4k blocks of geth trie state in the pathdb journal.
promotebakes that journal into the golden image and every per-testrestorehands it back, so geth loads it into memory on every boot. Accounts deployed in those final blocks are then served from RAM for the whole suite: in our runs one EEST account class read 23x faster than the same class on an identically-hosted generated-state baseline, purely from deploy recency. Draining the journal into the disk layer (and compacting) between pre-run and promote removes the bias.Usage
Unset (the default) the runner behaves exactly as before.
Write-up with the full investigation and before/after numbers: https://cperezz.github.io/articles/state-db-perf-divergence/state-db-perf-report.html