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

Immediately send request concurrency stat from activator on scale from zero #2659

Closed
greghaynes opened this issue Dec 6, 2018 · 2 comments
Closed
Assignees
Labels
area/autoscale kind/bug Categorizes issue or PR as related to a bug.
Milestone

Comments

@greghaynes
Copy link
Contributor

When we moved to purely metrics driven scaling we introduced a delay in the scale-from-zero case where autoscaler only scrapes concurrent requests and converges the deployment. In the scale from zero case where the activator handles an incoming request we should immediately send to the autoscaler.

@knative-prow-robot knative-prow-robot added area/autoscale kind/feature Well-understood/specified features, ready for coding. labels Dec 6, 2018
@josephburnett
Copy link
Contributor

Something like this and this

@josephburnett josephburnett added kind/bug Categorizes issue or PR as related to a bug. and removed kind/feature Well-understood/specified features, ready for coding. labels Dec 26, 2018
@josephburnett josephburnett added this to the Serving 0.4 milestone Jan 9, 2019
greghaynes pushed a commit to greghaynes/serving that referenced this issue Jan 21, 2019
Upon receiving a request the activator should immediately send a stat to
the autoscaler in the scale from zero case - this reduces cold-start
time.

Closes knative#2659
greghaynes pushed a commit to greghaynes/serving that referenced this issue Jan 22, 2019
Upon receiving a request the activator should immediately send a stat to
the autoscaler in the scale from zero case - this reduces cold-start
time.

Closes knative#2659
greghaynes pushed a commit to greghaynes/serving that referenced this issue Jan 22, 2019
Upon receiving a request the activator should immediately send a stat to
the autoscaler in the scale from zero case - this reduces cold-start
time.

Closes knative#2659
greghaynes pushed a commit to greghaynes/serving that referenced this issue Jan 22, 2019
Upon receiving a request the activator should immediately send a stat to
the autoscaler in the scale from zero case - this reduces cold-start
time.

Closes knative#2659
@greghaynes
Copy link
Contributor Author

/assign greghaynes

greghaynes pushed a commit to greghaynes/serving that referenced this issue Jan 28, 2019
Upon receiving a request the activator should immediately send a stat to
the autoscaler in the scale from zero case - this reduces cold-start
time.

Closes knative#2659
knative-prow-robot pushed a commit that referenced this issue Jan 29, 2019
Upon receiving a request the activator should immediately send a stat to
the autoscaler in the scale from zero case - this reduces cold-start
time.

Closes #2659
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
area/autoscale kind/bug Categorizes issue or PR as related to a bug.
Projects
None yet
Development

No branches or pull requests

4 participants