Commit Graph

3 Commits

Author SHA1 Message Date
Andrew Sy Kim f00505bddc priority & fairness: support dynamically configuring work estimator max seats
Max seats from prioriy & fairness work estimator is now min(0.15 x
nominalCL, nominalCL/handSize)

'Max seats' calculated by work estimator is currently hard coded to 10.
When using lower values for --max-requests-inflight, a single
LIST request taking up 10 seats could end up using all if not most seats in
the priority level. This change updates the default work estimator
config such that 'max seats' is at most 10% of the
maximum concurrency limit for a priority level, with an upper limit of 10.
This ensures seats taken from LIST request is proportional to the total
available seats.

Signed-off-by: Andrew Sy Kim <andrewsy@google.com>

Kubernetes-commit: d3ef2d4fe95c3ef7b1c606ad01be1183659da391
2023-04-26 11:13:14 +00:00
jupblb 16f776a534 Switch initial/final seats type to uint64
Kubernetes-commit: 3c46482eb09d7343e0f98a930a9aaa158237e278
2022-07-28 10:48:40 +02:00
jupblb 738a050cda Introduce config for API Priority and Fairness
Linked all the default values with a single config structure.

Kubernetes-commit: 1c594e7e01a899807431c806cd11c1d27c885c9c
2022-07-20 11:33:45 +02:00