Improve storage lookup performance #7336
Merged
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
anderseknert
force-pushed
the
ast-store-poc
branch
from
b062332 to
a7a48de
Compare
johanfylling
approved these changes
Feb 5, 2025
| return &storage.Error{ | ||
| Code: storage.NotFoundErr, | ||
| Message: message, | ||
| } |
There was a problem hiding this comment.
This is a performance optimization?
There was a problem hiding this comment.
ok, saw the benchtest 👍
There was a problem hiding this comment.
Yeah, it's quite insidious how much Sprintf can cost even in simple cases!
| sb.Grow(l) | ||
| for i := range p { | ||
| sb.WriteByte('/') | ||
| sb.WriteString(url.PathEscape(p[i])) |
There was a problem hiding this comment.
If this causes escaping, I suppose the only ill effect is that the above sb.Grow(l) wasn't enough and sb needs to grow further?
Since you're interested in performance in this PR, I assume you've confirmed this is better or equal in performance compared to the old method for the general/common case. How about the worst case, where sb needs to grow further (maybe more than once)?
Some code I had left from the weekend, where I looked into the cost of querying the store during eval, and if that could be improved. The code here is mostly benchmarks that I found useful for that purpose, but also includes a few optimizations of code that is on the hot path for these lookups and as such has a high impact — as the benchmarks demonstrate. The small change in eval.go is the first use of the new feature of object.Map added some days ago, where a map function returning a nil key means skipping that key/value pair. Signed-off-by: Anders Eknert <[email protected]>
anderseknert
force-pushed
the
ast-store-poc
branch
from
a7a48de to
efab512
Compare
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.
Some code I had left from the weekend, where I looked into the cost of querying the store during eval, and if that could be improved. The code here is mostly benchmarks that I found useful for that purpose, but also includes a few optimizations of code that is on the hot path for these lookups and as such has a high impact — as the benchmarks demonstrate.
The small change in eval.go is the first use of the new feature of object.Map added some days ago, where a map function returning a nil key means skipping that key/value pair.