Skip to content

Integrate kafka capacity planner with buffers #133

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: main
Choose a base branch
from

Conversation

ramsrivatsa
Copy link
Contributor

No description provided.

@ramsrivatsa ramsrivatsa self-assigned this Apr 18, 2025
@ramsrivatsa ramsrivatsa force-pushed the ramsrivatsak/integrate-kafka-with-buffers branch from 73ac91d to 773b600 Compare April 18, 2025 17:38
@ramsrivatsa ramsrivatsa force-pushed the ramsrivatsak/integrate-kafka-with-buffers branch from 773b600 to c65d9c0 Compare April 18, 2025 17:41
desired={
# Amount of compute buffer that we need to reserve in addition to
# cpu_headroom_target that is reserved on a per instance basis
"compute": Buffer(ratio=1.5, components=[BufferComponent.compute]),
Copy link
Contributor

Choose a reason for hiding this comment

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

ratio=2.5 here for tier0 and tier1 (to target 40% CPU utilization max on instance). It can be lower(1.5) for tier=2 and tier=3

# cpu_headroom_target that is reserved on a per instance basis
"compute": Buffer(ratio=1.5, components=[BufferComponent.compute]),
# This makes sure we use only 25% of the available storage
"storage": Buffer(ratio=4.0, components=[BufferComponent.storage]),
Copy link
Contributor

Choose a reason for hiding this comment

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

this can be 2.5X for storage (we are ok using upto 40% of disk)

"background": Buffer(
ratio=2.0,
components=[
BufferComponent.cpu,
Copy link
Contributor

Choose a reason for hiding this comment

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

how is background cpu buffer different from the compute buffer above?

core_count=needed_cores,
target_shape=instance,
reference_shape=reference_shape,
needed_cores = int(capacity_requirement.cpu_cores.mid)
Copy link
Contributor

Choose a reason for hiding this comment

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

just using mid cpu cores for needed cores felt not very accurate, hence I had the logic that you have removed. What is rationale for using the cpu.mid here instead?

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.

2 participants