refactor(storage): use dynamic trait for storage interfaces #846
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
Signed-off-by: Runji Wang <[email protected]>
Signed-off-by: Runji Wang <[email protected]>
Signed-off-by: Runji Wang <[email protected]>
Signed-off-by: Runji Wang <[email protected]>
Signed-off-by: Runji Wang <[email protected]>
we replace the async iterator trait by a boxed stream to avoid memory allocation on each call Signed-off-by: Runji Wang <[email protected]>
|
Weakly -1 The current dynamic traits have too many limitations, making them seem like unfinished products. And pass generic parameters are not really very painful. |
|
also objected by @MrCroxx |
The |
|
Given that the storage part of the system runs in batches (i.e., we always fetch as many as 1024 rows if possible), I don't think using a dynamic trait would cause a large regression. However, I don't see the code is simplified significantly without static dispatch, except that the |
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Outdated suggestions cannot be applied.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.
Signed-off-by: Runji Wang [email protected]
Dynamic async traits are not so bad for performance, as long as they are not called frequently. The advantage is that they are less invasive as they do not require generic parameters everywhere, and they allow for extensible plugins in the future.
This PR uses
#[async_trait]to refactor storage traits into dyn traits. It removes<S: Storage>from executors. The only performance critical part is theTxnIterator::next_batchwhich is called frequently in scanning. We refactor it into aBoxStreamto eliminate boxing the future on each call.Performance comparison on TPCH