Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

http: simplify sending header #33200

Closed
wants to merge 1 commit into from

Conversation

ronag
Copy link
Member

@ronag ronag commented May 1, 2020

unshifting into an empty array is the same as creating a new array.

Checklist
  • make -j4 test (UNIX), or vcbuild test (Windows) passes
  • tests and/or benchmarks are included
  • documentation is changed or added
  • commit message follows commit guidelines

unshifting into an empty array is the same
as creating a new array.
@ronag ronag requested a review from mscdex May 1, 2020 15:18
@nodejs-github-bot nodejs-github-bot added the http Issues or PRs related to the http subsystem. label May 1, 2020
@nodejs-github-bot
Copy link
Collaborator

Copy link
Member

@juanarbol juanarbol left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

LGTM

@ronag
Copy link
Member Author

ronag commented May 1, 2020

@ronag ronag added the author ready PRs that have at least one approval, no pending requests for changes, and a CI started. label May 1, 2020
@himself65
Copy link
Member

I notice that benchmark CI has running for over 21 hours. any problems or it works as expected?

@ronag
Copy link
Member Author

ronag commented May 2, 2020

I notice that benchmark CI has running for over 21 hours. any problems or it works as expected?

It does that sometimes... I'll restart it

https://ci.nodejs.org/view/Node.js%20benchmark/job/benchmark-node-micro-benchmarks/587/

@BridgeAR
Copy link
Member

BridgeAR commented May 3, 2020

@himself65 @ronag the issue is mainly the configuration of the benchmarks. Our default is set to run for five seconds per configuration for http benchmarks. This sums up easily (e.g., let's say a 2 * 3 * 2 matrix with five seconds each for 30 runs * 2 [before & after] = 12 * 60 * 5 seconds = 60 minutes).

It is easy to calculate how long these benchmarks have to run to finish:

  • chunked: 60 minutes
  • cluster.js: 60 minutes
  • end-vs-write-end.js: 120 minutes
  • headers: 20 minutes
  • incoming-headers: 10 minutes
  • set-header: 15 minutes
  • simple: 240 minutes

This has a total runtime of ~9 hours. This is an immense runtime for simple micro benchmarks. I suggest to drop as many configurations from the matrix as possible to reduce the runtime as well as reducing the duration from 5 to e.g., 3 seconds. Just the latter would reduce the runtime by ~3.5 hours.

The reason why it shows up as e.g., 21 hours runtime is that the job has to wait for former benchmark runs to finish before it starts. This is done to reduce the overall load for our systems, since we have limited hardware resources.

@ronag
Copy link
Member Author

ronag commented May 3, 2020

Landed in d799c0f

@ronag ronag closed this May 3, 2020
ronag added a commit that referenced this pull request May 3, 2020
unshifting into an empty array is the same
as creating a new array.

PR-URL: #33200
Reviewed-By: Juan José Arboleda <[email protected]>
Reviewed-By: James M Snell <[email protected]>
Reviewed-By: Zeyu Yang <[email protected]>
Reviewed-By: Luigi Pinca <[email protected]>
Reviewed-By: Ruben Bridgewater <[email protected]>
targos pushed a commit that referenced this pull request May 4, 2020
unshifting into an empty array is the same
as creating a new array.

PR-URL: #33200
Reviewed-By: Juan José Arboleda <[email protected]>
Reviewed-By: James M Snell <[email protected]>
Reviewed-By: Zeyu Yang <[email protected]>
Reviewed-By: Luigi Pinca <[email protected]>
Reviewed-By: Ruben Bridgewater <[email protected]>
@targos targos mentioned this pull request May 4, 2020
targos pushed a commit that referenced this pull request May 7, 2020
unshifting into an empty array is the same
as creating a new array.

PR-URL: #33200
Reviewed-By: Juan José Arboleda <[email protected]>
Reviewed-By: James M Snell <[email protected]>
Reviewed-By: Zeyu Yang <[email protected]>
Reviewed-By: Luigi Pinca <[email protected]>
Reviewed-By: Ruben Bridgewater <[email protected]>
targos pushed a commit that referenced this pull request May 7, 2020
unshifting into an empty array is the same
as creating a new array.

PR-URL: #33200
Reviewed-By: Juan José Arboleda <[email protected]>
Reviewed-By: James M Snell <[email protected]>
Reviewed-By: Zeyu Yang <[email protected]>
Reviewed-By: Luigi Pinca <[email protected]>
Reviewed-By: Ruben Bridgewater <[email protected]>
targos pushed a commit that referenced this pull request May 13, 2020
unshifting into an empty array is the same
as creating a new array.

PR-URL: #33200
Reviewed-By: Juan José Arboleda <[email protected]>
Reviewed-By: James M Snell <[email protected]>
Reviewed-By: Zeyu Yang <[email protected]>
Reviewed-By: Luigi Pinca <[email protected]>
Reviewed-By: Ruben Bridgewater <[email protected]>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
author ready PRs that have at least one approval, no pending requests for changes, and a CI started. http Issues or PRs related to the http subsystem.
Projects
None yet
Development

Successfully merging this pull request may close these issues.

8 participants