fix: lazy loading to improve startup time slows down large programs #35335
Conversation
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
This PR is a bit speculative; we think that the lazy property getters we
inject to avoid having to read all files of the entire `aws-cdk-lib`
library upon module loading (so to speed up
time-to-executing-first-statement) is adding significant overhead for
programs with a large number of statements.
Roughly:
```ts
// This avoids having to read the './core' module at startup, only do it
// if and when 'AspectPriority' gets accessed
Object.defineProperty(exports, "AspectPriority", { get: () => require("./core").AspectPriority });
```
However, the result of this is that every property access requires
running one or more getters (one for every level of indirecting file).
This may have a noticeable effect, on the order of ~1us per property
access.
To cut down on that, we will do the lazy getters differently: after
the initial lazy load, we overwrite the getter with the actual value.
That way subsequent accesses can bypass the getters and be fast:
```ts
// Only execute getter on first access
Object.defineProperty(exports, "AspectPriority", { get: () => {
const value = require("./core").AspectPriority;
// Overwrite getter with real value once we have it
Object.defineProperty(exports, "AspectPriority", { value });
return value;
});
```
This should achieve both goals of fast initial load and high throughput
for subsequent accesses.
aws-cdk-automation
previously requested changes
Aug 27, 2025
rix0rrr
added
pr-linter/exempt-test
aws-cdk-automation
dismissed
their stale review
✅ Updated pull request passes all PRLinter validations. Dismissing previous PRLinter review.
|
Confirmed that it has the desired effect on a large CDK app inside Amazon |
rix0rrr
changed the title
rix0rrr
changed the title
rix0rrr
changed the title
aws-cdk-automation
added
the
pr/needs-maintainer-review
kumvprat
approved these changes
Aug 28, 2025
|
Thank you for contributing! Your pull request will be updated from main and then merged automatically (do not update manually, and be sure to allow changes to be pushed to your fork). |
kumvprat
removed
the
pr-linter/exempt-test
|
Thank you for contributing! Your pull request will be updated from main and then merged automatically (do not update manually, and be sure to allow changes to be pushed to your fork). |
|
Comments on closed issues and PRs are hard for our team to see. |
Sign up for free
to subscribe to this conversation on GitHub.
Already have an account?
Sign in.
3 participants
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.
The lazy property getters we inject to avoid having to read all files of the entire
aws-cdk-liblibrary upon module loading (so to speed uptime-to-executing-first-statement) is adding significant overhead for programs with a large number of statements.
Roughly:
However, the result of this is that every property access requires running one or more getters (one for every level of indirecting file). This may have a noticeable effect, on the order of ~1us per property access.
To cut down on that, we will do the lazy getters differently: after the initial lazy load, we overwrite the getter with the actual value. That way subsequent accesses can bypass the getters and be fast:
This should achieve both goals of fast initial load and high throughput for subsequent accesses.
By submitting this pull request, I confirm that my contribution is made under the terms of the Apache-2.0 license