-
Notifications
You must be signed in to change notification settings - Fork 171
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
gui: moving window average and max for tile primary metrics #3166
base: main
Are you sure you want to change the base?
Conversation
ffaf01a
to
8e51a5d
Compare
174b0dd
to
456d757
Compare
904f651
to
61cad04
Compare
@@ -17,7 +17,6 @@ runs: | |||
run: | | |||
set -x | |||
sudo src/util/shmem/fd_shmem_cfg fini || true | |||
sudo src/util/shmem/fd_shmem_cfg reset || true |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Can you move all the CI changes out to another PR if they are still needed?
@@ -586,21 +586,21 @@ potential underflow. | |||
| Field | Type | Description | |||
|---------------------|---------------------|------------ | |||
| next_leader_slot | `number\|null` | The next leader slot | | |||
| tile_primary_metric | `TilePrimaryMetric` | Per-tile-type primary metrics. Some of these are point-in-time values (P), and some are aggregated since the end of the previous leader slot (A) | | |||
| tile_primary_metric | `TilePrimaryMetric` | Per-tile-type primary metrics. Some of these are point-in-time values (P), and some are moving window averages (W) | |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Mention that it's 1s moving average
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Also is there a separate part of the docs clarifying that it's the max when querying a historical slot?
Currently live values are computed from the end of the prior leader slot. Change to a 1-second moving average. Also change queries for our slots to return max within slot, so this would tell us for example the peak number of QUIC connections or peak fullness of the pack buffer that we observed within a slot.