Skip to content

Add initial-inflight-factor option to vespa-feed-client to mitigate the slow start #34247

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

Open
wants to merge 1 commit into
base: master
Choose a base branch
from

Conversation

yjtakamabe
Copy link
Contributor

Is your feature request related to a problem? Please describe.

The feed client experiences a slow start because the default initial target inflight value is too low in our environment, causing a 2-3 minute delay in the time it takes to reach optimal throughput.

Describe the solution you'd like

I would like an option called "initial-inflight-factor" that allows us to configure a higher initial target inflight.
This would help the feed client ramp up more quickly and avoid performance bottlenecks at startup.

    --initial-inflight-factor <arg>       Multiplier for minInflight to
                                          determine the initial
                                          targetInflight
$ vespa-feed-client \
  --initial-inflight-factor 64 \
  --file /path/to/json/file \
  --endpoint https://container-endpoint:443/

Describe alternatives you've considered

One alternative we considered was setting the number of "connections" to 512 in order to mitigate the slow start.
However, this approach consumes significant machine resources.

$ vespa-feed-client \
  --connections 512 \
  --file /path/to/json/file \
  --endpoint https://container-endpoint:443/

Additional context

In our environment, we set the initial-inflight-factor to 64.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

1 participant