Give both sides the same weather - #24
Merged
Merged
Conversation
The comparison script timed all of one call and then all of the other, and this is a laptop. Run to run, DuckDB's Arrow figure moved between 15 and 30 milliseconds and its pandas figure between 52 and 146, which is not DuckDB varying by three times. It is whichever call went second getting whatever the machine was doing by then, and fastest of five is not enough runs to filter that out when something else on the box is sitting at one whole core. The runs alternate now, ours then theirs then ours again, nine of each, and both keep their own fastest. That matters more than the count does: a noisy window now lands on both sides of the comparison rather than on one of them. Two consecutive runs of the changed script agree to within a tenth on every ratio in it, where the old one disagreed by two times on half of them. Also a decimal on a figure below ten milliseconds, because DuckDB's execute is a third of one and the table was printing it as 0 ms, which reads as a measurement of nothing rather than as the fastest thing in the table.
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
tools/versus_duckdb.pytimed all of one call and then all of the other, and this is a laptop. Run to run, DuckDB's Arrow figure moved between 15 and 30 milliseconds and its pandas figure between 52 and 146, which is not DuckDB varying by three times. It is whichever call went second getting whatever the machine was doing by then, and fastest of five is not enough runs to filter that out when something else on the box is sitting at one whole core.The runs alternate now, ours then theirs then ours again, nine of each, and both keep their own fastest. That matters more than the count does: a noisy window lands on both sides of the comparison rather than on one of them. Two consecutive runs of the changed script agree to within a tenth on every ratio in it, where the old one disagreed by two times on half of them.
Also a decimal on a figure below ten milliseconds, because DuckDB's
executeis a third of one and the table was printing it as0 ms, which reads as a measurement of nothing rather than as the fastest thing in the table.No library code changes, so the suite is untouched at 597 passed, 5 skipped.