request bare metal kubernetes cluster for PEAKS project

[knarayan]

Please fill out the details below to file a request for access to the CNCF Community Infrastructure Lab. Please note that access is targeted to people working on specific open source projects; this is not designed just to get your feet wet. The most important answer is the URL of the project you'll be working with.

If you're looking to learn Kubernetes and related technologies, please try out Killercoda or Play With Kubernetes instead.

First and Last Name

Krishnasuri Narayanam

Email

knaraya3@in.ibm.com

Company/Organization

IBM Research, India

Job Title

Senior Research Engineer

Project Title (i.e., a summary of what do you want to do, not what is the name of the open source project you're working with)

Evaluate the amount of savings in power consumption to execute a workload by using PEAKS kubernetes scheduler-plugin vs. the default kubernetes scheduler

Briefly describe the project (i.e., what are the details of what you're planning to do with these servers?)

We want to create a three node bare-metal k8s cluster. Then we want to run workload using the default k8s scheduler and measure the power consumption (using node utilization metrics collected from Metrics Server and energy consumption metrics collected using Kepler). Later we want to run the same workload using PEAKS k8s scheduler-plugin to schedule the workload pod on the cluster nodes and measure the power consumption. The amount of power consumption using PEAKS k8s scheduler-plugin should be less (leading to saving in power consumption). PEAKS plugin achieves this saving in power consumption as it tries to choose more power efficient nodes for pod placement at any time.

Is the code that you're going to run 100% open source? If so, what is the URL or URLs where it is located? What is your association with that project?

Yes, the code that we are going to run is 100% open source. URL: https://github.com/sustainable-computing-io/peaks/ and https://github.com/sustainable-computing-io/scheduler-plugins/tree/master/pkg/peaks

What kind of machines and how many do you expect to use (see: https://deploy.equinix.com/product/bare-metal/servers/)?

2 x m3.small
1 x m3.large

What operating system and networking are you planning to use?

Prefer RHEL 9 or CentOS stream 9 but open to other alternatives

Any other relevant details we should know about?

This issue is tracked in PEAKS: sustainable-computing-io/peaks#42

[idvoretskyi]

Ping @caniszczyk for approval

[idvoretskyi]

@knarayan apologies for the delay. Still valid?

[knarayan]

@knarayan apologies for the delay. Still valid?

Yes, @idvoretskyi. This request is still valid.

[idvoretskyi]

@knarayan thanks.

A related question - would the shared tenant for the Sustainable Computing projects, eg. Kepler - #253 work for you, or you'd like to have them independent?

[knarayan]

Hi @idvoretskyi, I request you to allocate an independent one.

[idvoretskyi]

@knarayan done, a project is created and invite is sent.

Please beware of the CIL usage policies - https://github.com/cncf/cluster?tab=readme-ov-file#usage-guidelines, specifically about the fair policy consumption - https://contribute.cncf.io/resources/newsletters/2024-02-21-projectnewsletter/#new-policy-on-resource-usage

[knarayan]

Thank you @idvoretskyi. Can you please add @husky-parul and @fali007 also to this project?

[idvoretskyi]

@knarayan I'll need the emails please.

[knarayan]

Hi @idvoretskyi , please find the emails here: parsingh@redhat.com and felixpv007@gmail.com

[idvoretskyi]

@knarayan invites sent!

[knarayan]

@knarayan invites sent!

Thank you @idvoretskyi