From ca6bc7f792fae8858cb638866d33eb0cb7a02073 Mon Sep 17 00:00:00 2001
From: Jiaxin Shan <seedjeffwan@gmail.com>
Date: Mon, 4 Sep 2023 22:48:24 -0700
Subject: [PATCH 01/16] KEP-4176: Static Policy to spread hyperthreads across
 physical CPUs

Co-authored-by: Lingyan Yin <yin.387@osu.edu>
Co-authored-by: Zewei Ding <horace.d@outlook.com>
Co-authored-by: Shengjie Xue <3150104939@zju.edu.cn>
---
 .../README.md                                 | 879 ++++++++++++++++++
 .../cpu-ordering.png                          | Bin 0 -> 16996 bytes
 .../kep.yaml                                  |  42 +
 3 files changed, 921 insertions(+)
 create mode 100644 keps/sig-node/4176-cpumanager-spread-cpus-preferred-policy/README.md
 create mode 100644 keps/sig-node/4176-cpumanager-spread-cpus-preferred-policy/cpu-ordering.png
 create mode 100644 keps/sig-node/4176-cpumanager-spread-cpus-preferred-policy/kep.yaml

diff --git a/keps/sig-node/4176-cpumanager-spread-cpus-preferred-policy/README.md b/keps/sig-node/4176-cpumanager-spread-cpus-preferred-policy/README.md
new file mode 100644
index 00000000000..eb7a955ac4d
--- /dev/null
+++ b/keps/sig-node/4176-cpumanager-spread-cpus-preferred-policy/README.md
@@ -0,0 +1,879 @@
+<!--
+**Note:** When your KEP is complete, all of these comment blocks should be removed.
+
+To get started with this template:
+
+- [ ] **Pick a hosting SIG.**
+  Make sure that the problem space is something the SIG is interested in taking
+  up. KEPs should not be checked in without a sponsoring SIG.
+- [ ] **Create an issue in kubernetes/enhancements**
+  When filing an enhancement tracking issue, please make sure to complete all
+  fields in that template. One of the fields asks for a link to the KEP. You
+  can leave that blank until this KEP is filed, and then go back to the
+  enhancement and add the link.
+- [ ] **Make a copy of this template directory.**
+  Copy this template into the owning SIG's directory and name it
+  `NNNN-short-descriptive-title`, where `NNNN` is the issue number (with no
+  leading-zero padding) assigned to your enhancement above.
+- [ ] **Fill out as much of the kep.yaml file as you can.**
+  At minimum, you should fill in the "Title", "Authors", "Owning-sig",
+  "Status", and date-related fields.
+- [ ] **Fill out this file as best you can.**
+  At minimum, you should fill in the "Summary" and "Motivation" sections.
+  These should be easy if you've preflighted the idea of the KEP with the
+  appropriate SIG(s).
+- [ ] **Create a PR for this KEP.**
+  Assign it to people in the SIG who are sponsoring this process.
+- [ ] **Merge early and iterate.**
+  Avoid getting hung up on specific details and instead aim to get the goals of
+  the KEP clarified and merged quickly. The best way to do this is to just
+  start with the high-level sections and fill out details incrementally in
+  subsequent PRs.
+
+Just because a KEP is merged does not mean it is complete or approved. Any KEP
+marked as `provisional` is a working document and subject to change. You can
+denote sections that are under active debate as follows:
+
+```
+<<[UNRESOLVED optional short context or usernames ]>>
+Stuff that is being argued.
+<<[/UNRESOLVED]>>
+```
+
+When editing KEPS, aim for tightly-scoped, single-topic PRs to keep discussions
+focused. If you disagree with what is already in a document, open a new PR
+with suggested changes.
+
+One KEP corresponds to one "feature" or "enhancement" for its whole lifecycle.
+You do not need a new KEP to move from beta to GA, for example. If
+new details emerge that belong in the KEP, edit the KEP. Once a feature has become
+"implemented", major changes should get new KEPs.
+
+The canonical place for the latest set of instructions (and the likely source
+of this file) is [here](/keps/NNNN-kep-template/README.md).
+
+**Note:** Any PRs to move a KEP to `implementable`, or significant changes once
+it is marked `implementable`, must be approved by each of the KEP approvers.
+If none of those approvers are still appropriate, then changes to that list
+should be approved by the remaining approvers and/or the owning SIG (or
+SIG Architecture for cross-cutting KEPs).
+-->
+
+# KEP-4176: A new Static Policy to spread Hyperthreads across physical CPUs to better utilize CPU Cache
+
+<!--
+This is the title of your KEP. Keep it short, simple, and descriptive. A good
+title can help communicate what the KEP is and should be considered as part of
+any review.
+-->
+
+<!--
+A table of contents is helpful for quickly jumping to sections of a KEP and for
+highlighting any additional information provided beyond the standard KEP
+template.
+
+Ensure the TOC is wrapped with
+  <code>&lt;!-- toc --&rt;&lt;!-- /toc --&rt;</code>
+tags, and then generate with `hack/update-toc.sh`.
+-->
+
+<!-- toc -->
+- [Release Signoff Checklist](#release-signoff-checklist)
+- [Summary](#summary)
+- [Motivation](#motivation)
+  - [Goals](#goals)
+  - [Non-Goals](#non-goals)
+- [Proposal](#proposal)
+  - [User Stories (Optional)](#user-stories-optional)
+    - [Story 1 Bytedance Database Performance Optimization](#story-1-bytedance-database-performance-optimization)
+    - [Story 2](#story-2)
+  - [Notes/Constraints/Caveats (Optional)](#notesconstraintscaveats-optional)
+  - [Risks and Mitigations](#risks-and-mitigations)
+- [Design Details](#design-details)
+  - [Test Plan](#test-plan)
+      - [Prerequisite testing updates](#prerequisite-testing-updates)
+      - [Unit tests](#unit-tests)
+      - [Integration tests](#integration-tests)
+      - [e2e tests](#e2e-tests)
+  - [Graduation Criteria](#graduation-criteria)
+  - [Upgrade / Downgrade Strategy](#upgrade--downgrade-strategy)
+  - [Version Skew Strategy](#version-skew-strategy)
+- [Production Readiness Review Questionnaire](#production-readiness-review-questionnaire)
+  - [Feature Enablement and Rollback](#feature-enablement-and-rollback)
+  - [Rollout, Upgrade and Rollback Planning](#rollout-upgrade-and-rollback-planning)
+  - [Monitoring Requirements](#monitoring-requirements)
+  - [Dependencies](#dependencies)
+  - [Scalability](#scalability)
+  - [Troubleshooting](#troubleshooting)
+- [Implementation History](#implementation-history)
+- [Drawbacks](#drawbacks)
+- [Alternatives](#alternatives)
+- [Infrastructure Needed (Optional)](#infrastructure-needed-optional)
+<!-- /toc -->
+
+## Release Signoff Checklist
+
+<!--
+**ACTION REQUIRED:** In order to merge code into a release, there must be an
+issue in [kubernetes/enhancements] referencing this KEP and targeting a release
+milestone **before the [Enhancement Freeze](https://git.k8s.io/sig-release/releases)
+of the targeted release**.
+
+For enhancements that make changes to code or processes/procedures in core
+Kubernetes—i.e., [kubernetes/kubernetes], we require the following Release
+Signoff checklist to be completed.
+
+Check these off as they are completed for the Release Team to track. These
+checklist items _must_ be updated for the enhancement to be released.
+-->
+
+Items marked with (R) are required *prior to targeting to a milestone / release*.
+
+- [ ] (R) Enhancement issue in release milestone, which links to KEP dir in [kubernetes/enhancements] (not the initial KEP PR)
+- [ ] (R) KEP approvers have approved the KEP status as `implementable`
+- [ ] (R) Design details are appropriately documented
+- [ ] (R) Test plan is in place, giving consideration to SIG Architecture and SIG Testing input (including test refactors)
+  - [ ] e2e Tests for all Beta API Operations (endpoints)
+  - [ ] (R) Ensure GA e2e tests meet requirements for [Conformance Tests](https://github.com/kubernetes/community/blob/master/contributors/devel/sig-architecture/conformance-tests.md) 
+  - [ ] (R) Minimum Two Week Window for GA e2e tests to prove flake free
+- [ ] (R) Graduation criteria is in place
+  - [ ] (R) [all GA Endpoints](https://github.com/kubernetes/community/pull/1806) must be hit by [Conformance Tests](https://github.com/kubernetes/community/blob/master/contributors/devel/sig-architecture/conformance-tests.md) 
+- [ ] (R) Production readiness review completed
+- [ ] (R) Production readiness review approved
+- [ ] "Implementation History" section is up-to-date for milestone
+- [ ] User-facing documentation has been created in [kubernetes/website], for publication to [kubernetes.io]
+- [ ] Supporting documentation—e.g., additional design documents, links to mailing list discussions/SIG meetings, relevant PRs/issues, release notes
+
+<!--
+**Note:** This checklist is iterative and should be reviewed and updated every time this enhancement is being considered for a milestone.
+-->
+
+[kubernetes.io]: https://kubernetes.io/
+[kubernetes/enhancements]: https://git.k8s.io/enhancements
+[kubernetes/kubernetes]: https://git.k8s.io/kubernetes
+[kubernetes/website]: https://git.k8s.io/website
+
+## Summary
+
+<!--
+This section is incredibly important for producing high-quality, user-focused
+documentation such as release notes or a development roadmap. It should be
+possible to collect this information before implementation begins, in order to
+avoid requiring implementors to split their attention between writing release
+notes and implementing the feature itself. KEP editors and SIG Docs
+should help to ensure that the tone and content of the `Summary` section is
+useful for a wide audience.
+
+A good summary is probably at least a paragraph in length.
+
+Both in this section and below, follow the guidelines of the [documentation
+style guide]. In particular, wrap lines to a reasonable length, to make it
+easier for reviewers to cite specific portions, and to minimize diff churn on
+updates.
+
+[documentation style guide]: https://github.com/kubernetes/community/blob/master/contributors/guide/style-guide.md
+-->
+
+We propose this KEP to introduce a new CPU Manager Static Policy that spreads hyper threads across physical cores. In this policy, we changed the cpu assignment sorting algorithm to sort by socket and then directly cpus without taking physical cores into ordering. It seems like a noisy neighbour issue, but it's always true. We will explain the reason in the motivation section. This policy is useful for some applications which need to take advantage of CPU Cache.
+
+## Motivation
+
+<!--
+This section is for explicitly listing the motivation, goals, and non-goals of
+this KEP.  Describe why the change is important and the benefits to users. The
+motivation section can optionally provide links to [experience reports] to
+demonstrate the interest in a KEP within the wider Kubernetes community.
+
+[experience reports]: https://github.com/golang/go/wiki/ExperienceReports
+-->
+
+`full-pcpus-only` was introduced to resolve the problem that different containers sharing the same physical cores which leads noisy neighbor issue. `distribute-cpus-across-numa` distributed cpus across NUMA to make sure no single worker suffers from NUMA effects more than any other, improving the overall performance of these types of applications. These two and default behavior can not meet the requirement of some applications which need to take advantage of L2 Cache. In that case, we need to spread hyper threads across physical cores, while NUMA is not a important factor. In such cases, the assumption is that the single physical core with two applications won't be always busy. So we can take advantage of CPU Cache to improve the performance of these applications. Otherwise, it still have the noisy neighbor issue.
+
+
+### Goals
+
+<!--
+List the specific goals of the KEP. What is it trying to achieve? How will we
+know that this has succeeded?
+-->
+- Introduce a new CPU Manager Static Policy that spreads hyper threads across physical cores without considering NUMA.
+- Enhance application performance by taking advantage of L2 Cache.
+
+### Non-Goals
+
+<!--
+What is out of scope for this KEP? Listing non-goals helps to focus discussion
+and make progress.
+-->
+
+- This proposal does not aim to modify the existing CPU Manager Core Binding Policies. It focuses solely on introducing a new policy for spreading hyper threads across physical cores.
+- It does not address other resource allocation or management aspects within Kubernetes.
+
+
+## Proposal
+
+<!--
+This is where we get down to the specifics of what the proposal actually is.
+This should have enough detail that reviewers can understand exactly what
+you're proposing, but should not include things like API designs or
+implementation. What is the desired outcome and how do we measure success?.
+The "Design Details" section below is for the real
+nitty-gritty.
+-->
+
+We propose to add a new `CPUManager` policy option called `spread-physical-cpus-preferred` to the static CPUManager policy. When enabled, this will trigger the CPUManager to try to allocate CPUs across physical nodes as much as possible.
+
+### User Stories (Optional)
+
+<!--
+Detail the things that people will be able to do if this KEP is implemented.
+Include as much detail as possible so that people can understand the "how" of
+the system. The goal here is to make this feel real for users without getting
+bogged down.
+-->
+
+#### Story 1 Bytedance Database Performance Optimization
+
+We're running DB instances in Kubernetes and adopt default static policy in the past. While, we notice that the performance of DB instances is not stable. If an instance is under pressure, in original way, it was allocated two cpus from same physical core. However, an important pattern we notice is not always all instances are busy. After exploration, we find that the CPU cache is one bottleneck, once we spread hyper threads across physical cores, the busy instance can leverage more CPU cache and performance is improved a lot.
+
+#### Story 2
+
+### Notes/Constraints/Caveats (Optional)
+
+<!--
+What are the caveats to the proposal?
+What are some important details that didn't come across above?
+Go in to as much detail as necessary here.
+This might be a good place to talk about core concepts and how they relate.
+-->
+
+### Risks and Mitigations
+
+<!--
+What are the risks of this proposal, and how do we mitigate? Think broadly.
+For example, consider both security and how this will impact the larger
+Kubernetes ecosystem.
+
+How will security be reviewed, and by whom?
+
+How will UX be reviewed, and by whom?
+
+Consider including folks who also work outside the SIG or subproject.
+-->
+
+The risk associated with implementing this new proposal is minimal. It pertains only to a distinct policy option within the `CPUManager` and is safeguarded by the option's inherent security measures, in addition to the default deactivation of the `CPUManagerPolicyAlphaOptions` feature gate.
+
+| Risk                                             | Impact | Mitigation |
+| -------------------------------------------------| -------| ---------- |
+| Bugs in the implementation lead to kubelet crash | High   | Disable the policy option and restart the kubelet. The workload will run but with CPU packing semantics - like it was before this new policy option was added. |
+
+
+## Design Details
+
+<!--
+This section should contain enough information that the specifics of your
+change are understandable. This may include API specs (though not always
+required) or even code snippets. If there's any ambiguity about HOW your
+proposal will be implemented, this is the place to discuss them.
+-->
+
+Current default sorting order is `sockets`, `cores` and then `cpus`. Using machine with 2 sockets, 6 cores, 12 CPUs topology as an example, default cpu ordering is [0, 6, 2, 8, 4, 10 | 1, 7, 3, 9, 5, 11].  In that case, if cpu manager plans to allocated two cpus, [0, 6] will be picked up. However, they belong to same socket, same numa node and same physical core which can not meet our case.
+
+
+In order to meet our use case, we can change sorting algorithm to sort by `socket` and then directly `cpus` without taking physical cores into ordering. In that case, we get cpu sequence [0, 2, 4, 6, 8, 10 | 1, 3, 5, 7, 9, 11]. From the topology information, we know [0, 2] will be allocated for 2 cpu container and 0 and 2 are from same socket but different physical cores. 
+
+
+![](./cpu-ordering.png)
+
+
+
+### Test Plan
+
+<!--
+**Note:** *Not required until targeted at a release.*
+The goal is to ensure that we don't accept enhancements with inadequate testing.
+
+All code is expected to have adequate tests (eventually with coverage
+expectations). Please adhere to the [Kubernetes testing guidelines][testing-guidelines]
+when drafting this test plan.
+
+[testing-guidelines]: https://git.k8s.io/community/contributors/devel/sig-testing/testing.md
+-->
+
+[x] I/we understand the owners of the involved components may require updates to
+existing tests to make this code solid enough prior to committing the changes necessary
+to implement this enhancement.
+
+##### Prerequisite testing updates
+
+<!--
+Based on reviewers feedback describe what additional tests need to be added prior
+implementing this enhancement to ensure the enhancements have also solid foundations.
+-->
+
+##### Unit tests
+
+<!--
+In principle every added code should have complete unit test coverage, so providing
+the exact set of tests will not bring additional value.
+However, if complete unit test coverage is not possible, explain the reason of it
+together with explanation why this is acceptable.
+-->
+
+<!--
+Additionally, for Alpha try to enumerate the core package you will be touching
+to implement this enhancement and provide the current unit coverage for those
+in the form of:
+- <package>: <date> - <current test coverage>
+The data can be easily read from:
+https://testgrid.k8s.io/sig-testing-canaries#ci-kubernetes-coverage-unit
+
+This can inform certain test coverage improvements that we want to do before
+extending the production code to implement this enhancement.
+-->
+
+- `<package>`: `<date>` - `<test coverage>`
+
+##### Integration tests
+
+<!--
+Integration tests are contained in k8s.io/kubernetes/test/integration.
+Integration tests allow control of the configuration parameters used to start the binaries under test.
+This is different from e2e tests which do not allow configuration of parameters.
+Doing this allows testing non-default options and multiple different and potentially conflicting command line options.
+-->
+
+<!--
+This question should be filled when targeting a release.
+For Alpha, describe what tests will be added to ensure proper quality of the enhancement.
+
+For Beta and GA, add links to added tests together with links to k8s-triage for those tests:
+https://storage.googleapis.com/k8s-triage/index.html
+-->
+
+- <test>: <link to test coverage>
+
+##### e2e tests
+
+<!--
+This question should be filled when targeting a release.
+For Alpha, describe what tests will be added to ensure proper quality of the enhancement.
+
+For Beta and GA, add links to added tests together with links to k8s-triage for those tests:
+https://storage.googleapis.com/k8s-triage/index.html
+
+We expect no non-infra related flakes in the last month as a GA graduation criteria.
+-->
+
+- <test>: <link to test coverage>
+
+### Graduation Criteria
+
+<!--
+**Note:** *Not required until targeted at a release.*
+
+Define graduation milestones.
+
+These may be defined in terms of API maturity, [feature gate] graduations, or as
+something else. The KEP should keep this high-level with a focus on what
+signals will be looked at to determine graduation.
+
+Consider the following in developing the graduation criteria for this enhancement:
+- [Maturity levels (`alpha`, `beta`, `stable`)][maturity-levels]
+- [Feature gate][feature gate] lifecycle
+- [Deprecation policy][deprecation-policy]
+
+Clearly define what graduation means by either linking to the [API doc
+definition](https://kubernetes.io/docs/concepts/overview/kubernetes-api/#api-versioning)
+or by redefining what graduation means.
+
+In general we try to use the same stages (alpha, beta, GA), regardless of how the
+functionality is accessed.
+
+[feature gate]: https://git.k8s.io/community/contributors/devel/sig-architecture/feature-gates.md
+[maturity-levels]: https://git.k8s.io/community/contributors/devel/sig-architecture/api_changes.md#alpha-beta-and-stable-versions
+[deprecation-policy]: https://kubernetes.io/docs/reference/using-api/deprecation-policy/
+
+Below are some examples to consider, in addition to the aforementioned [maturity levels][maturity-levels].
+
+#### Alpha
+
+- Feature implemented behind a feature flag
+- Initial e2e tests completed and enabled
+
+#### Beta
+
+- Gather feedback from developers and surveys
+- Complete features A, B, C
+- Additional tests are in Testgrid and linked in KEP
+
+#### GA
+
+- N examples of real-world usage
+- N installs
+- More rigorous forms of testing—e.g., downgrade tests and scalability tests
+- Allowing time for feedback
+
+**Note:** Generally we also wait at least two releases between beta and
+GA/stable, because there's no opportunity for user feedback, or even bug reports,
+in back-to-back releases.
+
+**For non-optional features moving to GA, the graduation criteria must include
+[conformance tests].**
+
+[conformance tests]: https://git.k8s.io/community/contributors/devel/sig-architecture/conformance-tests.md
+
+#### Deprecation
+
+- Announce deprecation and support policy of the existing flag
+- Two versions passed since introducing the functionality that deprecates the flag (to address version skew)
+- Address feedback on usage/changed behavior, provided on GitHub issues
+- Deprecate the flag
+-->
+
+### Upgrade / Downgrade Strategy
+
+<!--
+If applicable, how will the component be upgraded and downgraded? Make sure
+this is in the test plan.
+
+Consider the following in developing an upgrade/downgrade strategy for this
+enhancement:
+- What changes (in invocations, configurations, API use, etc.) is an existing
+  cluster required to make on upgrade, in order to maintain previous behavior?
+- What changes (in invocations, configurations, API use, etc.) is an existing
+  cluster required to make on upgrade, in order to make use of the enhancement?
+-->
+
+We anticipate no repercussions. The new policy option is voluntary and operates independently from the current selections.
+
+### Version Skew Strategy
+
+No changes needed.
+<!--
+If applicable, how will the component handle version skew with other
+components? What are the guarantees? Make sure this is in the test plan.
+
+Consider the following in developing a version skew strategy for this
+enhancement:
+- Does this enhancement involve coordinating behavior in the control plane and nodes?
+- How does an n-3 kubelet or kube-proxy without this feature available behave when this feature is used?
+- How does an n-1 kube-controller-manager or kube-scheduler without this feature available behave when this feature is used?
+- Will any other components on the node change? For example, changes to CSI,
+  CRI or CNI may require updating that component before the kubelet.
+-->
+
+## Production Readiness Review Questionnaire
+
+<!--
+
+Production readiness reviews are intended to ensure that features merging into
+Kubernetes are observable, scalable and supportable; can be safely operated in
+production environments, and can be disabled or rolled back in the event they
+cause increased failures in production. See more in the PRR KEP at
+https://git.k8s.io/enhancements/keps/sig-architecture/1194-prod-readiness.
+
+The production readiness review questionnaire must be completed and approved
+for the KEP to move to `implementable` status and be included in the release.
+
+In some cases, the questions below should also have answers in `kep.yaml`. This
+is to enable automation to verify the presence of the review, and to reduce review
+burden and latency.
+
+The KEP must have a approver from the
+[`prod-readiness-approvers`](http://git.k8s.io/enhancements/OWNERS_ALIASES)
+team. Please reach out on the
+[#prod-readiness](https://kubernetes.slack.com/archives/CPNHUMN74) channel if
+you need any help or guidance.
+-->
+
+### Feature Enablement and Rollback
+
+<!--
+This section must be completed when targeting alpha to a release.
+-->
+
+###### How can this feature be enabled / disabled in a live cluster?
+
+<!--
+Pick one of these and delete the rest.
+
+Documentation is available on [feature gate lifecycle] and expectations, as
+well as the [existing list] of feature gates.
+
+[feature gate lifecycle]: https://git.k8s.io/community/contributors/devel/sig-architecture/feature-gates.md
+[existing list]: https://kubernetes.io/docs/reference/command-line-tools-reference/feature-gates/
+-->
+
+- [x] Feature gate (also fill in values in `kep.yaml`)
+  - Feature gate name: `CPUManagerPolicyAlphaOptions`
+  - Components depending on the feature gate: `kubelet`
+- [x] Change the kubelet configuration to set a CPUManager policy of static and a CPUManager policy option of `spread-physical-cpus-preferred`
+  - Will enabling / disabling the feature require downtime of the control plane? No
+  - Will enabling / disabling the feature require downtime or reprovisioning of a node? (Do not assume Dynamic Kubelet Config feature is enabled). Yes -- a kubelet restart is required.
+
+###### Does enabling the feature change any default behavior?
+
+<!--
+Any change of default behavior may be surprising to users or break existing
+automations, so be extremely careful here.
+-->
+No. In order to enable the feature, the user must explicitly set the `CPUManager` policy to static, enable `CPUManagerPolicyAlphaOptions` and the CPUManager policy option to `spread-physical-cpus-preferred`.
+
+
+###### Can the feature be disabled once it has been enabled (i.e. can we roll back the enablement)?
+
+<!--
+Describe the consequences on existing workloads (e.g., if this is a runtime
+feature, can it break the existing applications?).
+
+Feature gates are typically disabled by setting the flag to `false` and
+restarting the component. No other changes should be necessary to disable the
+feature.
+
+NOTE: Also set `disable-supported` to `true` or `false` in `kep.yaml`.
+-->
+
+For sure. The feature gate can be disabled in many ways.
+- Set `CPUManager` policy to `none`
+- Disable `CPUManagerPolicyAlphaOptions`
+- select other policy options instead of `spread-physical-cpus-preferred`
+
+Ongoing workloads will maintain their operations without disruption, while upcoming tasks will receive CPU allocations in accordance with the reinstated policy.
+
+###### What happens if we reenable the feature if it was previously rolled back?
+
+If we reactivate the feature after a rollback, the outcome remains unchanged. Current containers will retain their allocations, while newly created containers will be affected.
+
+###### Are there any tests for feature enablement/disablement?
+
+A dedicated e2e test will validate the preservation of the default behavior when the feature gate is turned off or when the feature is unused. This will be conducted through two distinct test scenarios.
+
+<!--
+The e2e framework does not currently support enabling or disabling feature
+gates. However, unit tests in each component dealing with managing data, created
+with and without the feature, are necessary. At the very least, think about
+conversion tests if API types are being modified.
+
+Additionally, for features that are introducing a new API field, unit tests that
+are exercising the `switch` of feature gate itself (what happens if I disable a
+feature gate after having objects written with the new field) are also critical.
+You can take a look at one potential example of such test in:
+https://github.com/kubernetes/kubernetes/pull/97058/files#diff-7826f7adbc1996a05ab52e3f5f02429e94b68ce6bce0dc534d1be636154fded3R246-R282
+-->
+
+### Rollout, Upgrade and Rollback Planning
+
+<!--
+This section must be completed when targeting beta to a release.
+-->
+We only target alpha for this release.
+
+###### How can a rollout or rollback fail? Can it impact already running workloads?
+
+<!--
+Try to be as paranoid as possible - e.g., what if some components will restart
+mid-rollout?
+
+Be sure to consider highly-available clusters, where, for example,
+feature flags will be enabled on some API servers and not others during the
+rollout. Similarly, consider large clusters and how enablement/disablement
+will rollout across nodes.
+-->
+
+In the worst case, if the logic has panic and kubelet crashes, the kubelet will restart and the workload will run with the default policy. Running workloads won't be affected.
+
+###### What specific metrics should inform a rollback?
+
+<!--
+What signals should users be paying attention to when the feature is young
+that might indicate a serious problem?
+-->
+
+I verify the correctness by checking the kubelet log and the CPU allocation of the workload. I have not added any metrics against this new feature.
+
+###### Were upgrade and rollback tested? Was the upgrade->downgrade->upgrade path tested?
+
+<!--
+Describe manual testing that was done and the outcomes.
+Longer term, we may want to require automated upgrade/rollback tests, but we
+are missing a bunch of machinery and tooling and can't do that now.
+-->
+We manually test it in our internal environment and it works. It's worth doing automated upgrade/rollback tests in the future.
+
+###### Is the rollout accompanied by any deprecations and/or removals of features, APIs, fields of API types, flags, etc.?
+
+<!--
+Even if applying deprecation policies, they may still surprise some users.
+-->
+No.
+
+### Monitoring Requirements
+
+<!--
+This section must be completed when targeting beta to a release.
+
+For GA, this section is required: approvers should be able to confirm the
+previous answers based on experience in the field.
+-->
+
+###### How can an operator determine if the feature is in use by workloads?
+
+<!--
+Ideally, this should be a metric. Operations against the Kubernetes API (e.g.,
+checking if there are objects with field X set) may be a last resort. Avoid
+logs or events for this purpose.
+-->
+
+Examine the kubelet configuration of a node to verify the existence of the feature gate and the utilization of the new policy option.
+
+###### How can someone using this feature know that it is working for their instance?
+
+<!--
+For instance, if this is a pod-related feature, it should be possible to determine if the feature is functioning properly
+for each individual pod.
+Pick one more of these and delete the rest.
+Please describe all items visible to end users below with sufficient detail so that they can verify correct enablement
+and operation of this feature.
+Recall that end users cannot usually observe component logs or access metrics.
+-->
+
+- [ ] Events
+  - Event Reason: 
+- [ ] API .status
+  - Condition name: 
+  - Other field: 
+- [ ] Other (treat as last resort)
+  - Details:
+
+###### What are the reasonable SLOs (Service Level Objectives) for the enhancement?
+
+<!--
+This is your opportunity to define what "normal" quality of service looks like
+for a feature.
+
+It's impossible to provide comprehensive guidance, but at the very
+high level (needs more precise definitions) those may be things like:
+  - per-day percentage of API calls finishing with 5XX errors <= 1%
+  - 99% percentile over day of absolute value from (job creation time minus expected
+    job creation time) for cron job <= 10%
+  - 99.9% of /health requests per day finish with 200 code
+
+These goals will help you determine what you need to measure (SLIs) in the next
+question.
+-->
+
+Even CPU allocation algorithm is changed in this case, it won't cause any performance regression. So we don't need to define any SLOs for this feature.
+
+
+###### What are the SLIs (Service Level Indicators) an operator can use to determine the health of the service?
+
+<!--
+Pick one more of these and delete the rest.
+-->
+
+- [ ] Metrics
+  - Metric name:
+  - [Optional] Aggregation method:
+  - Components exposing the metric:
+- [ ] Other (treat as last resort)
+  - Details:
+
+###### Are there any missing metrics that would be useful to have to improve observability of this feature?
+
+<!--
+Describe the metrics themselves and the reasons why they weren't added (e.g., cost,
+implementation difficulties, etc.).
+-->
+
+N/A
+
+### Dependencies
+
+<!--
+This section must be completed when targeting beta to a release.
+-->
+
+N/A
+
+###### Does this feature depend on any specific services running in the cluster?
+
+<!--
+Think about both cluster-level services (e.g. metrics-server) as well
+as node-level agents (e.g. specific version of CRI). Focus on external or
+optional services that are needed. For example, if this feature depends on
+a cloud provider API, or upon an external software-defined storage or network
+control plane.
+
+For each of these, fill in the following—thinking about running existing user workloads
+and creating new ones, as well as about cluster-level services (e.g. DNS):
+  - [Dependency name]
+    - Usage description:
+      - Impact of its outage on the feature:
+      - Impact of its degraded performance or high-error rates on the feature:
+-->
+
+This applies to any machines with hyper-threading enabled.
+
+### Scalability
+
+<!--
+For alpha, this section is encouraged: reviewers should consider these questions
+and attempt to answer them.
+
+For beta, this section is required: reviewers must answer these questions.
+
+For GA, this section is required: approvers should be able to confirm the
+previous answers based on experience in the field.
+-->
+
+###### Will enabling / using this feature result in any new API calls?
+
+<!--
+Describe them, providing:
+  - API call type (e.g. PATCH pods)
+  - estimated throughput
+  - originating component(s) (e.g. Kubelet, Feature-X-controller)
+Focusing mostly on:
+  - components listing and/or watching resources they didn't before
+  - API calls that may be triggered by changes of some Kubernetes resources
+    (e.g. update of object X triggers new updates of object Y)
+  - periodic API calls to reconcile state (e.g. periodic fetching state,
+    heartbeats, leader election, etc.)
+-->
+No
+
+###### Will enabling / using this feature result in introducing new API types?
+
+<!--
+Describe them, providing:
+  - API type
+  - Supported number of objects per cluster
+  - Supported number of objects per namespace (for namespace-scoped objects)
+-->
+No
+
+###### Will enabling / using this feature result in any new calls to the cloud provider?
+
+<!--
+Describe them, providing:
+  - Which API(s):
+  - Estimated increase:
+-->
+No
+
+###### Will enabling / using this feature result in increasing size or count of the existing API objects?
+
+<!--
+Describe them, providing:
+  - API type(s):
+  - Estimated increase in size: (e.g., new annotation of size 32B)
+  - Estimated amount of new objects: (e.g., new Object X for every existing Pod)
+-->
+No
+
+###### Will enabling / using this feature result in increasing time taken by any operations covered by existing SLIs/SLOs?
+
+<!--
+Look at the [existing SLIs/SLOs].
+
+Think about adding additional work or introducing new steps in between
+(e.g. need to do X to start a container), etc. Please describe the details.
+
+[existing SLIs/SLOs]: https://git.k8s.io/community/sig-scalability/slos/slos.md#kubernetes-slisslos
+-->
+No
+
+###### Will enabling / using this feature result in non-negligible increase of resource usage (CPU, RAM, disk, IO, ...) in any components?
+
+<!--
+Things to keep in mind include: additional in-memory state, additional
+non-trivial computations, excessive access to disks (including increased log
+volume), significant amount of data sent and/or received over network, etc.
+This through this both in small and large cases, again with respect to the
+[supported limits].
+
+[supported limits]: https://git.k8s.io/community//sig-scalability/configs-and-limits/thresholds.md
+-->
+No
+
+###### Can enabling / using this feature result in resource exhaustion of some node resources (PIDs, sockets, inodes, etc.)?
+
+<!--
+Focus not just on happy cases, but primarily on more pathological cases
+(e.g. probes taking a minute instead of milliseconds, failed pods consuming resources, etc.).
+If any of the resources can be exhausted, how this is mitigated with the existing limits
+(e.g. pods per node) or new limits added by this KEP?
+
+Are there any tests that were run/should be run to understand performance characteristics better
+and validate the declared limits?
+-->
+No
+
+### Troubleshooting
+
+<!--
+This section must be completed when targeting beta to a release.
+
+For GA, this section is required: approvers should be able to confirm the
+previous answers based on experience in the field.
+
+The Troubleshooting section currently serves the `Playbook` role. We may consider
+splitting it into a dedicated `Playbook` document (potentially with some monitoring
+details). For now, we leave it here.
+-->
+
+###### How does this feature react if the API server and/or etcd is unavailable?
+N/A
+
+###### What are other known failure modes?
+N/A
+<!--
+For each of them, fill in the following information by copying the below template:
+  - [Failure mode brief description]
+    - Detection: How can it be detected via metrics? Stated another way:
+      how can an operator troubleshoot without logging into a master or worker node?
+    - Mitigations: What can be done to stop the bleeding, especially for already
+      running user workloads?
+    - Diagnostics: What are the useful log messages and their required logging
+      levels that could help debug the issue?
+      Not required until feature graduated to beta.
+    - Testing: Are there any tests for failure mode? If not, describe why.
+-->
+
+###### What steps should be taken if SLOs are not being met to determine the problem?
+
+## Implementation History
+
+<!--
+Major milestones in the lifecycle of a KEP should be tracked in this section.
+Major milestones might include:
+- the `Summary` and `Motivation` sections being merged, signaling SIG acceptance
+- the `Proposal` section being merged, signaling agreement on a proposed design
+- the date implementation started
+- the first Kubernetes release where an initial version of the KEP was available
+- the version of Kubernetes where the KEP graduated to general availability
+- when the KEP was retired or superseded
+-->
+
+## Drawbacks
+
+<!--
+Why should this KEP _not_ be implemented?
+-->
+
+## Alternatives
+
+<!--
+What other approaches did you consider, and why did you rule them out? These do
+not need to be as detailed as the proposal, but should include enough
+information to express the idea and why it was not acceptable.
+-->
+
+## Infrastructure Needed (Optional)
+
+<!--
+Use this section if you need things from the project/SIG. Examples include a
+new subproject, repos requested, or GitHub details. Listing these here allows a
+SIG to get the process for these resources started right away.
+-->
\ No newline at end of file
diff --git a/keps/sig-node/4176-cpumanager-spread-cpus-preferred-policy/cpu-ordering.png b/keps/sig-node/4176-cpumanager-spread-cpus-preferred-policy/cpu-ordering.png
new file mode 100644
index 0000000000000000000000000000000000000000..76bb1e06085620edb53ce2351413fa9e9063fc00
GIT binary patch
literal 16996
zcmdtJWmKEr+wTc9P-sgbNO37%D6Yj)iWe_VaVeVMZbe!w1X@aQXbJ8V2rk7^oIr62
zPH?9<!%zM*=hd8fG3S|i&Wn4ky^?k9-1olszT|U#6Q!xHNce#I0Tvb(p|X<P8!RmB
z#k)%dAMfsy#Q~v@SXhsw>}6#&m1SiaHQk)w**n=_VJSr=>fble?s<~=_H+EpR|JfV
ztjBDTzp>uDvct~trg=(0{2Cha%~Fwq+_*9FJ)sG?*0zv@rLER>VCA>$k{H~^LpEHG
zIsF#z7U$jTJ&^5<``T>14VETokUYj+ixBHj4)^n9JeGpnYtFD^GDfU>*5B^i8Jo5w
zlhe^*e<RGih3u8dI@{^fS{vPN-<Id~P^=tbz07{9zP^5R$S944Wja8r@&Jpm*XQUN
zt|iq+LOT`nP-5OMi>kO~Ul%=a=b8UR#o5Tt;=PZiPU7kTy!RqFguXIm`~B@iW}9W{
zp*&WBAmiykUs`8`7@;K%hWVpG-jM384EvFD>V|BpF=UNcuU{U*8PXo$%HutKYY4jq
z`__;Aq|~b@v{8O#W#j(QvoS^RneoRl#$EDUMFp7>+=Kg1&31|SU-x|e$i{J{C2J~8
zGyVwkYP)kC@1v{}zH0@CsKBe9d|Ujm7sMUpq3YyMT$Y}WhxvSG1aYWuRA~u3)1!I_
z{uN(g`1<C#K2sc5jHg*@Lff0o)<F`NFBg=h99+@_mJd3gd}!}finJ$|SX=M>z!j)K
zVN=O`Kagd_!0_e?G4g@jQ|t$UC490Zw8h7~1APqXya;}kUr{<sfw+bT-=i<g!IEW?
z!D1BKpI>&qru+?DVOM-K_<<My(VKoce>P&)&S##|kq>b!$*o+Y#YBxZ;?fA*WUZU1
zI?K*Kxb{~1SYjK;KM0hRH72Gj|3%8%yY^9@>+ok%2!0!0$b0OZirg?|MjCtxKN|nV
z#B6p-mk@w84z`P?Kp6Aq6)7ek=so!bjYt4vu$}SyCVi|2Pj5XVb!q8der&4MlxDgw
z86rj$9%k&roft<#D6ayDd&+#J=)vP8ne5Et^y4OYi0B8J>WMYe7pMmxlWgE{0;Sv|
zmX9Q1QHVP1-5CF6;68>ZO>nSN>i+t{T-JX0u(FR*odSkwHkIS~*Q+n9l*70uVTm<L
zx0O%o$bWlu+;d?lc%0~JGK(d)P{#N8;h(TibBuj@Yuv>Tmf3e&2Noh4zaKM8e9}7P
z`vIC-TmK{^kvbC}T`~?DSN!C7q1ikoD+NN@-f9L)OGeyoM{#$aTG>{RPQKXY8^@=d
z<S!v`M8baU4{!QCH!}%CS+1V1yrf$MEfLNe8nPZ`Y^M9Aw{ToM0BGCn*|lY|WU>L@
zy(8x{*pFc}caf7s0w3A3+frazqba(!w$z|kjUgCB@cP;{_WPmAkdP2rbbW;t*7nK!
zxZor2!H)zWx+1J}@R~B~hX;?7=<xTszdz0r3N(MzA3~TOC+Kp2zKsD8!bgm|B=aB{
zt7<58%NQr`J(W2R2eyqll>s5=PrM2$r~8d<(!=<V!$=o|>aci192XuNOU*6dV(#s=
z*#cyg@X5b3UqljEMolsg;*%c8hsNWxJW-YziC2{iSNcTtOETw?VTcY>TD*}p?+7&g
z<>1qdU)(F$BVkPn(%&hIK03e^3dqGtY_gvV(Z(tw;VBsFYW$2i9vQA-Y7h45_n9ww
zzfBDD9OF*MVXKmlcB%5;helOaay%#RYh#&wUP8u&bg1wv!JS{{GN2fb3G9p#qqALj
z?1CGyL?A|#1s!UqSWZ0iYQlEzC|@Ajm%gG2dCkW%N!o-@`%2(x0^bMCR~+n4th6lY
zp+j#(7wJAJ{bn^Hm0GAue$~&W&d$SD#=Of$&+bLS0K8OY`7GU&xgs0Q<f@FDLo#AC
zqCUbuV)I6HBP_ZoIx4#Ab;awVXr)|(Tw|MvUaiDFP7Ph@1mRy_t$E$ZQZzCZOkdBx
zf;=^2@lF6~3q2brV|XpenkJC+My4=7J$p8{NlRB(So=gLR*Sr7P^(mP4(_LS60)f%
z%q=V-jF=V^E}SN*0XrbaAcf7}-xTWU1Uqonyl)6MQjt<_*0_ENAEmXYrOpyb)021n
zW|G&e!(RNhsOmR+0Yza_@t%Hb!F;jIc*Tghql@r6dmVcr$Jt@bM2;hD1Uk|`YFtt;
zdz8K=$o=_iqF($HGsl+kcclqNaTv<E^ufF%%n{}-CLc35EeCc4!@Lk)DW^O;k9VZ$
zxassW<T47yb;T7u8a((-xJ+P>vfVdxfejO$AVNXvZfZ74gq^{4MPJAK7U9Iv1SuwQ
zb7#CU?~iiUcZQrSc&l-%)@>#>#N3wHim%JO`RB8+`Nipu=7^(@NsrN@SBR$$cD9s`
zVskE47L69IpDMK?KTX<|qLksLRwh=#%TCL>-HNgPsu;EwwpR(dY)I8iK@ZC33Ow*o
zR5wOYScp(KF%=@P!*}iU4*b<z&j~TJo%y0G-9FQzrMGFUX{f!^6591BMm%QbtF=m=
z&DT}xKEbbhr6r}GOP&}A8%b4-SFIUP7)}_XYrapPO{pRZE9)Ips%@s%Cd;atts|!E
zr);Klr%b99cjk6JoNjGPC=x69z=&fDs9ZZ_20~UfF`Dc9ZCVrj?Uj?2sDo9dm`{bw
zNpX^bhiM6E+GV3p@+IKa3Lj!Q>Bj^%YpOB?hw6LJiyn@#PTC+;68gCG{q;k({3g;M
z9<}_ChqFd=Ee-y4DzjU2{*uonM<k1+Abd3`sQ#}0s!YHT*3eXDgcQiH-k;oO=jftu
z=BEVGKjKdnY8;LWuZ0G!Sgm!OOs$Nr_U)@3&0%I%Z8wo8#RvWv=iQ+{=*`<NHhsPC
zcC4Epxjcvm&_41RI<*|KW+3Xsk;PHDClOj5N)pN$wtGJf*A{OA_s@sA52{2j=wIU#
z5}DBZg;{kjx;nfFw4k>T?3ivJYhP~X>R?cqlviRh>i%Zc-d#fLE;=IchHQ!wH(N}L
z3%)D=R#5{cO(qMCf)z&ll1OlRDQu^R`OS9L(Yfv|-CIiPMk`k+$Fo(l6|uuqO4I_f
z$dap>3b}l(5BpahKEr#~{v}W6jf3Up`0txa87uv`9r|^VGV9rsgYo5j7)bP|`K#S<
zX3{ZsH}+i(<{Z}C9i_|m*rjoo%h4AbV;g~zBF=<+xah3bXf*Rau!LB0V86%U)4}sm
zc6Xb18eh%yeD%uXxl71edLIwW$6rgcbK1Tgd<l88yDOR4RNeIM(^n3~-y%IeBfph1
zEd5=h>KuPetc)dGvMhcsJy*IpyUN1>pE09(YChdC1!`j68t9%`yk7cptu34%orln<
zN<-L9k7h4qqw|(WeMjHgCJ!b{EdNv;?^mmeH{7T)vP~Px1U34r|4QF+f9CGKc4NA&
z)0BO-)0E@OeH6PklyZwniZ|PQZO8igGtI*$3MYd|UNW9-gOwk4JK4^;S0hI@>Ncx3
zW>1cuWZioHE+=n&{Jh6tr1E#*#)VC%O}5QM(j_;#<(0cUDq4{XS>j~SWuV)zxw9En
zJe1h@)!9JSklAEZ7pR4*9=F{2QDeqg#aXV4tFu!kQevue9YhyAw|a83wIxak=9t;H
zhr28fs69=ib|`V^6(tp=fCT+ub?r+*C%2A4geQ#<CtD&YMA=T|%#4GSgQGX)u7Od+
zxbl{t7i`AYUVou1&oAOat<lpvX$RXs$3JsVZN@*)?5t({y8rs*{&ON7h8%|H;Fr{S
z!pFkzguR4wg*-(fgliE#wX~8eK6Wz}qaC`69<b+9uV0AYV%jRRt*((>NPi9MOrPK4
zl_!HeD+&5Dmeb<uAF{;##m|m=1WN>wgBR&i0fSq{N4{(4-4AyusHf+qiEegJFMmG%
z1N#_5Pm#vsYwEB;JLh}+1HDSBk5iuB(p&GhXtYu@xJ5AGJWuHj-(tI6t9n~sYdXi%
zbbUUU-WcAP`$uRu5iQzeb`-J0(AYTJ2$RGQ7{1}#UCdByQskvy2i=TF-4dKApKB-N
z*9@g+(d`oK)*lCKAMH{sQDph6-2A?(y2u^qyBB<ZJ2sy*J8}K`&|%?ZN%X=Q(d-#G
zGtDp;d}}h_IJBn%!URv=d^$hq&(G1&0BM}Rx_EOicIwzyyYK9M{}2za)0ja2hU@*$
zkMFFp_WfwEev@HgT!Iy-iYIS-2(`8iSMjgdOD36n2k74Wo!!!pKD>oMuqtdH&IZW@
zp15MkX=6wJ0TjgaY0}IV_xwo|U~Ug=>&BFC_--tqlP!nNW#>UzA*y5KR{=K^@$mst
zJLCl$clCsrje+tzH8m{GyD>f%c7#0^;BJI{cTwM6SXemUaQ^cN_TsmD|L0icuP<2#
z@nBe3!kNl)GCDrkI~fnFO>}+tWJ&PD|7eqjun_ms002}keLgT1W;1E3#K87Ru<f;E
zy6yw&dY6mHh%25vkmshr<IMe)J1)fS*2n$TsDeYY@U$CY`PLsdW~Ywx&#i>*K!w3}
z^U$=!ddgJ`|Bdb7dFi=PYt8A!xUqkRd!E`T3nMmx(n}nY`1gSS?n84;oUKRG!(Jo?
zVffo$1E-`e$}!IHJ^9wiiYZj4Y&}Kr6~|Pof2G)$8CH;J><$0LO}y>d{c8wWDfGsE
za_nH8YgvlLi1J$O!t3~@&e|s@#~Ef}TX_uykp0|dE_q7d1!{tM4X*spaw#OX>22)9
zpdJ!(YovfDrM288Y;a~jXw`qpCA^7&H%NWKx%o4e5qs713`C<<UPk_GDE2~lO#u{f
zI#Z#Sm4<gAR<S0?g-rkz2uXp@X3Z&^a{|wrRZk}|^_WrAIBEp*3lj&w&TcSCYae3P
zOKqhMe136>f!S$Vb)8QY9V8{VX}3vAlhM)hBh8|MfU)^=@jpdkZhR&0dp`gBr5Z@!
z(`L*)FCOf0Y;4jVXbnC*N#u*OjX5YtKGoNU$>_Ip6$N3+ONurQzx57Ls$`#|^s{N0
z^#+L)BNQj8fHrz<43U{IJ8!#i?#MbVA2vUs2Bsxf;chkj>r6TmYLu5(XDs%O2{ZIC
z5JiBJ0$QPlytFDi5<>1R^9^~+AG-$R{JUCuKt=;#f#y-vaMPI=R0P(5yQugh7BL}5
z`7`3Mi_Bqs%T;K31TVwPQT8i%6?lT%Q+2`Psp6*-kngphv;$ZifRAV#B#<aodMkK>
zd_(tpd(G$-Ne$9YI7pNETP}C2>CnD9!z_#;Nu<d)Zb|e6CFB6FveVCsZ6L?(15i$X
zbUgC6XQK!%{pgR>+1ly-dQLi)6m7?2+Zsv3HILv`F89*t(U!zV#TF0%Phom+{1;vO
zXZpMLdrBYE2uv%Mx+;JCkP`95v7IwN;6eqD-@RGSXXP}mR`0o~)IaJqTT`u*1F!8!
z;@j{6bu;5E6(S@YA#0$ENniTQCzloitqed9zbPm^&21?H4B<BWaa7!|5d(isq<pc5
zQa~XR_lk1Ranf`q#^*`~GFERDYCjiFh16VG^|XH1+x^-sNK+?MCwj40^ZHWPmtMAG
zEm*-U=81eNfMxi@dzre2lyi=fA$vOhXQ?@W-Yk6^Y0?s(^TA{7#BvrS>ya6Xh)J*G
zBq?{_V!&M7c~XijxoEp+<$MJ{9WY`B+l`6_3jHa60M*^Dx5jUy<eIVULQ3L?&gjq>
z;970T=iX#J_WgMI>?E@wfSDCk#z|(Hd-G8@Z+2EQ0O2H*_#Plmy6@^?%i}i@fWu0>
z6K`@o;NUH)bkTw-iQF?Hi1%y+;p+LIBR8>=#pCvwy?k_VMj$;&jQT`5uH`=(i)YK(
zOzL_iuHI-r`;onE&AMN;vtQJ1s@#`cmsxa$R&S^q@l*~Lz33~TES?#wL>6~Dt2L;1
z!Vf0&rAhzsU~PT)v(y^CDUr2-hV_Yu<ULlGa$U7I@%cwYdZoGZf<3&>Um9!XBu@&=
zrpuHv>%Sb@^jNKSm8@+Zb~|x-Zdol!2o()&RXs!NkL5Gyb?=q*W@#AJCO#Wh^tCow
ztFzU$&v%gUthkt2%Y~#Q@xCJzJk0;wYyQUHk32mr0K|MIIU6tC91G)%EAXd67mC~?
z<}%ltX@-&?g=tZ><}kqGURX9No~6koGY*zxCyEfU3>AGj%*K~^QG_<<Lv}E4_eVC&
zh3-2Rl5t(?wphIzGwr4=if71IVD+hLx8?c$f_>y)H%&)CY;3p)bP?S&fP>bH-(>8i
zif)uo4Kr+rF@n8f8%om~J?jHSzCjfD*e6LPZ9R&sse|-pk@EdnNK_TvP0+>Q@NVz?
z<NmGW7Uni;OQ43hNr{3jhi;RUgy33|<OV?Z63l0F#AWSnSAPTCa1sVnALxYaMfkeL
zB2ms!K60i19&Z-D$AEXR4$8+*|F;Pz|9b?*|5Z<64rZ;e^}70rnZlgfS%mLUmqzY`
zXz6GzF6F%CDdgf5Cb3cXP=Y8h>MIc8>!B8Ydw`i#O87U+U<8cE;m3<A#$@|vr)3+W
zdND1Sc+8JT^CmU9W{PZf_$XXoXVo!*o_dZ37+-k=zr2fsMDo>eXSIJ)m5Omv0}>oS
z_Y`>)t}<^NENb7GmYf+si!71RSg(J*OR>$Az1<6DciD@%z<jcci`)Sm3+Vm0+wgo^
zfIN-gZLqt*Tf$Tk=%yh*q@TIR*36em5y<7o!+B~TP~yjbg{JCNOG8+VfZ=9x05m(3
zQ*q?`n^8;}@WhNRvGViSzX!*WH-!1e7GN-iJhWbUDAuQ>zt>wKyfQjBK(q68asWTQ
z?okFmmvOay=O5&je93!JDE{ejpFyL-zemr!l>Oy5mT#Hlz?aTYNbGp*eijw*o3E2v
zN`N3f10BQ9#I|M`53OTWxw%HT3iHJbQ#L)Xm;HazweU77;7?#>_8fQnVAMV^Kv^S}
zuVtc3zg8nh^MVnJ!4UkM3%m(W<7|!k&rv7G*8Ax$K0Ntcn*Lc;Aj^z6c|jk|gy#&)
zhh`>v703U~{QvKq(7+MDq-uP-P(oY}0DFyW1(s6Gkq16Y<&faNmqk1f35nZ{38Idz
zP4sQ@Y`RY$c7RaMs#S0al%^(HZt&3aP(AL5Tt4fv%Lcz=+u!JB5kt4m?i+r&+nszo
zfYTukUbzW*z>+O|86H(JVXL`z7!_GBNf#(Y?ff<#I|Z4rEVa)OCn|BBK^w?w87vap
z-8VS+a!8IsE?*8BEGL9Q$g;J)xPhgfW?+<Gv;bHtPD$!sG~T34?~-CAP4q6OG>eCs
zn9=N~uhhrK<yXG(<7*I1-IVcxoG!xbm_qlZsQxq+P`cL@k+Jcaq*pQ|w;j+K-R=A~
z6?|eB^>W|&lq;Y6Edp1N9;detltDDWN61m>^q<lU&8&CR)Va(ErW}|O9C_lz<#<Pb
zRe%QT2%)?b*eV2;n%F6`$PR>|Y<GUz{MFrT8YQvD^aK;e%b9a5%=2Wfi_&(qiXc9#
zpH_Xp7$wCQoK-|=k+$$=h5Wv;&DH4YHOiypn`_`vgTUXBNEIKu#apUaL;(BzlaX&z
z)ur`Ip3nUjNv^c%x-`$rV$XIDmm+=R%|ugsMSxt3;^n2<3$Dk}4ep_bRf2b?t<qX-
zFcpBu#uvEoRpd)swLSV___;8#q;x=u9{tevF4guBJWlbxoQB-8O#tjh(1<pWQ(R)H
z)bfx<-Y6JA|N2C>nC<rS@k3O5k>sbIErGLQ2hy{Mm0(S-w7aZ?g|7_Ya~`GW7C_Or
zY|z469ITkWR5xWSK!P6KF&>pOsf!z+y;2t=#uG9<Yh+JlaD^PuZReuoXhePD6snj8
z1vBf6DgQoiY~fgOMtiWc`Ns!LzoB+a&~N%{T5ll4Z0+MYj7^GI?FoC+$iCBomy&DT
z$WKQOmj~2MNU>RBN!mo;Pg$ezmbAZo2J|5|J*wHwSivp6<^z^9UDY)0&;t7F&a95m
z86ww4>i_*>Ecp%K?8vPu|2H(F<ZmZXA`D<V0=X;@EhK@6^xxd5&MI@q4cZqX8G_f*
zFha|S9IbPrXGfpQZ~IzgE~;Yx@f_y8#6I%xUhhi@Ffkqx;Xt7K+_4Hj$P3$UuV{Zi
z$+bA2>A2!HztPS3YiHd>9Rn`>FN{szR?fUCzBU83LcTXu1Tu%8?cRA<$W?Uz`HJ81
z$LHAkxW`VR+M;tawG}t!z<HhH2&Ae)Y#J|lJ&ZQ_^q-lQ2|OOQmp}<3uPpws>fX%e
zD9QuYW0-Erxv~Gif!JO{k&1oJ(PyCCmADv@SWL{z81NIVJT-#$fP7V~e8Sr~`G4aM
z^g&3^RfbC~-PIq5jtk|XXb7w<riI#s_FCcC8`UbWA$Lo5{hSFiPjfAs+x!_(xvst8
z49}7ap34jI{MX}y@m0GR`)|pwEm&{HrtE>m`f@KbNrhvvE(RmbuLkUr3K9-D{K0|{
z;|Hw=N>t7)G<z>9(cm@VktTM(##y!te6be5I3vy+y&^g#aD}1Aib!5xPEtVO%53JE
zX*&c1maVrREmMCz=c}6{U1y$%L_OrXl~y*3+>C=yt;85=8u$+^ApDR)0*7f(cO$7@
z#Md?sQVyA9<gXGRf-+GBK+oaawoT7b_81K4tVx>g99Q8-yv8pSSHeFnO=Ch`s=W$|
z|B_S<iN>^2g_|)jpb5%jxmLFk|E#woO^A9cyDFc_n|nWFbo!Wa*!{%rYCZ*GX<rUe
z@H9%9pb4`P`Cj$9nb~xObH};(5U?oAv}yH)pdAqKZ+K=5iN-3PItLcp9L_oEi9C;d
zO&3IHub`WP94Sbud@x_9k6Zhn%<=ygKqOlOdjG%^oQ%|ML-F^U&p)APve&agu`MJg
zj#p0;nVIID)^QtKBc;g?A~Y^4!?e#mfx!kgHXgyps|5_pjaS;)VNCPyZ)Y$cF=lX?
zo%ce<PY&Lo&E&%-vxUUDEg?;m2DJm)3$a&SC~e^7m4Q$}#;1?RzPed-`<Bb0^xz=T
znV?}S|G?2EGIi)_R15jF>v1e*T?^FRNV`oGj7ub97TfaV#EK^Ouu($jEbdHt>L<Ee
zm}I$llpre~<S?4z$-*~<uR0Wdb@Hte#98|e9|y<zLy~BBFxPp|4ov8qlygE-%kE}X
z!g8@RCIZWNt)o1lrpEKlL1sZUxBH@+H_uuaC56j;MgEW#-UcK1o!so~@O}4^8N6p-
zZc=PhBt(H>lX~`@L4u~OdTR!2!>WHEQ;S!>xh~&;zW(VNvTZB{Cq?IGxjj!AotgI1
z9J87q2>h;^6*j2^rO4I3*jt~av!+LUYKsJ%VY!MLJLOyNA6QtEj}HWK)6j1tT%p%T
zBf)`#R+6*fo<rrXuHWmVnUq{4zbc-2!KSIX#HIss9NNl1&1sz@N&>(ngAc30y1sEa
ztL-xg1{?}^iz_uIJxr3qnKUJ#rDzUtN2^uF$8<VC$gkQ#gJFO%z_2FPoqm~UxB^|V
zP**J+)U}%$=>-*3BF;?kk`U6tq0E{>h;}FC^SCe2c8J8CNiNT@J2<ZH0bHzpRnAIt
zUOP=75`O!;cTP6S1BRK_#^Pc-&sIFV-gX-NVx+~N?j%_V)x<F(XfuPx5$W)z{bFoF
zT-B5etz)@-pRD#`d~F^SJRthUOP<pueT%{lCj7G;%$Jqc(4R}iLBM%N6gN#tA+wp}
z4ikCEg{~pHsCEl;b(}^yx5}JVR9mn{+Qp^HB){+5Yn&(rtc7h30uT54PA?3?0E_Cf
z#uV)CD{&ju!Z@qUvAxb3Dc2G9Z<m{1jXDzc@=JeTqj`3jse|bZ%JCZ5Vdh_pyYKoa
zS#X&u{GLpj0Zuy9n7E!YVZFT-5)P(xJy^tn9!$m^jss=R4k8<0^I!!O7*dhHP3?!H
zMp3LlY;y&<OKqm<UDHI3;@;m<Qk-5d1&HMOV_Xfu+>r&y`<OEPEgd3|Hwr*7ElB^<
zpOcD0xo#YcB%AbxX}`6)U2Nn3xHknX&JThjjJN-Qg&G8H)~d`p(51gPKVSS&d@Gl(
zofWy?{B#p3Ml@ABq_Db{gJv{lAj<29(}%x^*yoPf)Wgxc4k&eIcs6AFPAGJL>A2(S
z?9qu;ME;?44c_(<H!!er2n=mgxxS=MJU3pxRq8+&`vi&{Qn9+2ku<X}HHVE=nnJym
zk^LyfCh>5v1Sg;_(sd-^FmMrCK*`0Y{u?(j44M3epci0qXk0f>34Cq`5;P*I+?mA`
z*x7h!b&6aCh;g+3;2ed;VAOO|bj7?+em9GV3pI!`F%H3apvikJTD5*tLOWNHF;dt|
zvcT(s=)6!@E$bF<A+|n3lABe+;M%2b-D}0-F9>;tt?`1~TQyxX+SU^z#P{h4xIo0`
zc~Ub_;4qPflNb>`bvo!+cv>de9U3$Q-Sl#Ka0W8)KK&Fd;;Ao+f8@JVr7P_n_^7pr
zGfPJL?Xlpm8$OYp&8%W=PV|oQ4E1wNe&Nxy*YI~!-fq0b$88>JkFa?x)6YsYl?e*E
zrU8p;MZNy4xz~e03LPAWO6`Pa6c09^mx>T;K4Z{IRB2nd*B%k~@ZL{+xY?!~N=(F3
z7<y8KoI7C#a(P_5V6p5kGzUC6{qyy4!s^nY`Tbz_iA>iKq6VgEX(P(S<z|lapjdVi
zQHA3k*M-U28nAZd>|K6Y!C(_~zh;N!<g4BetY{jlOT319e&p{jtghU9URNS^q3Vwt
zi7;svCQqSpr~O?QwW^SKK1_l9=$=!Wkq}+I?Cu5e0PA)mUyO`XCLM*i+rR5PcrJ=}
zoo;52==I|l!7p#OewR%^9Bv49U2<Q4of7kW{89VcE@Xrw<k%4<tHnQEwbNEGOQ+4V
zbWdw@J5^FIbiaLZX;@=(kIsZ55pnY6y`Pg#s7tw@@jgSz#rK=W>_SUH0j*Pi-w1-`
zTg$r#r6?&Ja{Um@Pr1}uA!d{pk#AaW-Vp@<y7Rh*h`pok#H!M4)$uYi137&BGvU)U
z*R|fNzxPNWT&84n*$(oEaw{zM@%e+CdWKj|shDOQxuK6^3J^|$001Zb!}2oMmP9Le
zErtwNbuES`^0V8Y*k6{z4p?G>@OQ%u{(6-%k!7r4h2?u=yhV*83$`d{1sb`m=*{l?
zWHS&k!xQ&3-Mc8irhkt7e8bToFkM0Nb3d|xlA)yGy2-TQ@K$tV0zQI#*TX;}MLt%l
zFE!o7lS*;zB{$5-p@tjQLj*T-Q2m<t@U;1R7W3^HvZ#0Bvl}!<O3Oxs7xrW$P1{!f
zcrNlKvg?3d3xudE3V2%z{26X_y`^=n@Lg|})i9I~!C@&MC)P5_&U`_`y-aB?@rb=R
z$W7e}&7j_W&7Oq|`)QkPcKqUE_IInGc;z_kXy;>5fWZLh%~$u$k;r8{(iWaYzff1f
z`EI)o{*1ALe%G#xW)ux%RfFJ>k~veTOUh;|i@Q-L=by7kiJtXrQl`+w7g}?5e6AMr
zVwSjv?5HD@K72X*1(4y}7W1ydC4Xr54azZ_f=N%j<fd+c@nSU$T|6J$clt;1mbS9@
z=Qx9UUguDwpbC#u<Jh~)JWir5k#2q#O&7`fpPYI;)o30Va769LwF*#FMsnv&K`7D{
zX7Mm0;d_>GfME*peqBMFb#T@Lz_*5ZL*b=v^F2-wb<gNV_<<?KK$)ij&Scg^z;cSo
z8^Ti%9_5!Ip`-OG(^{z_98p8?PKZu85E)M+4MWc`ZyzjLcV_wL;URBA6$XC9fiLZa
zYB@p*WfoFY6Ns$m@V$(iYNL6g@k22P&Q<l)mje-Tr%rHb-eU7K`9Ao_?&3;q9DeFu
z@Z{rh_(Gvs06jF-y&+XrVSu_@ui6ImuBtb_mC{33+d?pTcASR3E9XjVfli%WvDYE&
z6jIx>_)n0U7)zDqd^!qMm~h;;MVZD}=Iy~qx7tMh?TuwF;JTdl(Q+CPg^VspbDbY1
zq8Jy1vjRnx&ZMa=z!y|y=SxE!3Au6N5fgd+0bk|8CH+<z3adwN^7bUMyHGzO581OV
z!7$lEl?TBFBH)~5%}vnfN0;M0)B9Gmr;9%#hq#Z{kHYu%@f|z%Qh&n-5>ZW&R<F;)
zv?@PMx|kSHp<t(%a<)1)Yt@kO8XH05MhPp#9tr>aeGWU4<ZXb+06>d9jeZ@bb^h&(
zxbd$wMG3`$7$f6S@Ye5@$AQ-h52fp9CIaR*D?^zJsy=5$>xDj{cqF&)rwdlZFE~T%
z90^xdTqx;WFs|Z7$C&Wr-~>UM#HlEX{Vv}HX(8MXkVCZoR2uB5JI*j?udxC!=!E}q
z@x{uAOiKVo_qnwzG(HGXgqAdWW@ZBBI%?rsy?B9M6U*u0WX4Cg79Cr@%UOP@260P4
z?vRNa&;+@ULgaN$Bfh-gj*YHatd-{!y*v*HNO$j=u9%Wa^gYzh8tQn)`#ggM+>xgj
zXcsKbpZ9ozwR_3<_g^K$F-wIh;s<BuMDF<ES>I?FkM76WStUA3SBvFWCZdox28OQM
zHR1|<@45s&^HZEgT>!%KYIgFc$_#HbX>CO_?tM)W3@Mx$4MiZoPwX%I($>{45RjMD
zqqs$7ddhx?Twx-<c^T-m@|%t+3xKDFGu=JZn_UwbxJBq1###TW^7(M`bEvq6-f8aC
z*3OJ9Gz8HIZ<4kVtgwq*T|ZGQ&N30?D{j1SC_>j)m$tIeN1&EOQzomWGROa_nuepX
zxQOXB%E}IKW;gNN^yEq)`V$Jp4D_XY6%7gYQ$cEykKtX;0+sf@Vx|8UZ2z^rL?5b<
znB_&Y(!#>N_g0ewk}cYQ>b?$HmoLxk)NLjG3mpGfVbK5Cywz+jz7D2i4W}+$_Y!C&
zU!}QMT~`A)>wH~s;DNJ<K^<u-?zl5s0tv-bKzDlEF+amkY!<$Hf|7)dlgk%kPr*UW
zeUKfR+YhK3l)A|##x(FRyWOELPZVqA(iAl0-pbt>etFqSk;R_@%~1J9goDi8jlQF@
z2)hLQ8DY+why-4)@su9kn9L~pxOsA6aJd+LG^|YIx!Q%@v4@AU90T+ch5@54;dyjq
z$YMByh^PTQ6CYLFj7SQVsD?`R&>hYll6Qs*{ax1E5l?ciQe>0D;Y>KI{{I8IWgL({
z?dQC`$YEEkHX}9x21+g+FnfYXER5p|JFINFzoBEYU!dqvbEx)$w7C?*<I-j&^B?rJ
zu>$gt%;=rZkC2&Z0{@hc)BuD=N(@V}A<wrj1ldTrGf`MH&_E;FfSgn=rdcpWmKkSN
z3dptM2)8_gaDK?iSJ2S22|Yd0ZFy@+9P}ck&}Fu6aml#u&+6y|^WQY$Lz{oX(*-7;
z<bc~=ha|-TPk+7-M@*soT<dwja1{_>-V!jo#QQAJ9B<L$aoAOQ2@?Fy>mv;p?BoN`
zMh(YGy=^ZXNn3djiXFl|y3U}mrKqS7z|i#k!mHXuh`HUzH$pMGB=nXom4@73MWDR!
z!LfX^71!d8mWNHXMJ_mj9%vs%n%s}v`BU{a;6tT~hwpKfQbJoYDIfk_N9aD~J)EPF
zbu-2CXg!^!*zpLa?c>PAett4bz#^}_bKII@qoP@45Um9|s)fpgZjN{W)<R83F%adm
zGQhcUg};o`Jf=Cwa7~<Cf{gV&KwPqp4epQOO*e6-{f%*7hGC)WTZj*?>Ed90F~aLQ
zLG*N08GCO0rh|Sg?MM;R>2@F5z%fn8S#}@72K5eg7<m!-a?m@KlGl5&5%!9rkvT|D
zXIh0owR_nsAkP5iE+uuDt6j_T_u6{m9KFFDH`W1$rLw0^dUb<la<)A8z1>Pl*RLXV
zf#(h2&+R1*9RZFw^tIy!B^-Yn{rAAk|Fw?nzwfV^5NxM#6DbTS)G9t?cABu055HZ+
zEa1<uH*hsD`U$l^j--#wjDo}<pcMmfN2aSdMdW`PATM#ARs_ktZ+c-wHb*)~HW00+
z@JBvEE=eA+3})JFCYGYgpl{CRG)p@k#>_{)hh{L$G3-zTa%9s5{@bevsmW~nOw7nP
zSf=z`$<NnAH9UB^fKuixrvR|{!%~`88s}6Th;Wyi(N~yIjFa=1b7KFN^mhj*r01pi
zg>`bcdwiRblKJKKkv8gRtR+}<;{&P*^A_c<Rf+s;F&49UJE#enuBJ<X&|EVyok;^N
z-pT`xvlP(Kg4@jkXky8KIHf-U?I#m<m%$R^el<Txr<ILfiy#wJ_xWlVlV1DEo$4ZW
z4{ErWt;XB}GUc_zZMp-Y`DWeI!vAUMI!a;VuwtK-L}K|_c{tuBPq+P%SOZp$y(?)l
zXP7hmGQcijc1*S{)8_nN;tTW>;E7MA%Kz0<8;rz<MUwA8fL`k1aSJMb4&evQJS`1@
zOdm$>W1szUurjpQW|-4CSNj)bk`&-qJjfgk>3(-`pp>%1JT+eU(qz7_Y=-UGHu-^p
z(SK(Cp9Q)9%3e_j|I;>*_QbzWp#8ZGFU9lx{_4PSr|T?ySzsJ?4~hh7p#9%_iG2De
z_N$6v--NvQ`DXwX`my_nApw+Hd7t0{Z}I&<+9z`d5)K)kTb|KY&YKzd1`3K9!~~+2
zFnO3%Op2X+wY!d!y#Qc5AFh&pIO#LA!T`n3JZ|J`c)p}yZsVa7&K)aWgJ!;2MI`{Q
zM+OsJ81EDyfAt+s?(lbjR$kB>wN|~I2O)_sYlXxGnS2JLCSkXNsHZ3?)Ep4OVmVU}
zlu2B8eY>Ry*iM7X1Ht$!P??InK|QV7(<vdHy_=kpRDw0U#g})|mcNp@zwS1f@k~jL
zXu@m#UJq#iYKAkIpuMW-&`X&h@>oCAvBrM7u&2U}`)8s!Wn3#SLoZ>uU*9qYvg7^Q
zq?=LmP85N+_>}DCZ*LPK0(gfg6vFnzFEZN`*sVDrecs9tSf;mLxpG^~d25xpoA@Ph
zrxInK7`IYycZn&)oa0|YJQg)~b3qL?AfYk4+j3pc?yOA_E(sUw<B#|X!PJ4Qn-gBD
zVaxCdxm)1%9%c<=i(*8mQ(t=>$6)5}d9(%0-$|BSAujV*<pm5gl8Hl6XAi=Tpr@@A
z*R6KMZ@(y{C*;e`HjTwrnlf<1#E>ntrGw|53e%(Ox>~K>YMVVM$2%t9zxuNv0P-1Z
zW1Fj2Z&4~*_w++>v4r0WC>@;sFav5eGIVoLgL=z6NH{ojdDl^v0kd@L=UYi@1PfuX
z5D9HQ#0&R^aP4pD!!nY<LwYB5!&ZQuMA`a9hDMg}S#;SE*<4H)3gywKF|f!2LBaMl
zh4sg)rVS*!<C*5V&Y*Hmjn%zy=hD;f`!=^hT$@PqA*-NiH&rFnwN~vzJCT#>(W*pg
z`d!1~BLzrjFD-rf>YokcW_S{o-f7|B$H3J2N_ppXsMFfoM_=I-Ct~#I7(#&U<Z(0}
zKMsTYyQU!UwGhWqPTY^Z{$lNyJ7d8w@CLpd<;MOjoI)5vMu|LOWhc6KJle}<8B2u3
z3Q~u;y?5^zOiBRd3Ly!vKU!Ndq+l2KE=})emrs&2euRj`Z$C?tJ9Y(kir{S90KMy~
z=h&7SmuBE6Ow#Xd<*V!Y8-&~GqXubvV`pzn3Kjivq}-E>Wa^BQ%Z8+hS>DO_F`wG6
z7nII@j;=N|7WT0j86iUEo4rlOJC2~=t>V9wNlQc_bZ0s~k4z-$)~g-f6b<+%xT%Aw
zD&4ZPWU~<>yyPHiDQeUc;jn$1#4LB}<-AWvUyf3)?H&8K{w2O^KB%ul)ZMmgq(JGx
zO9aA`bL5?U=XzaP(5b!5w40R)h3L|8pvCWQ%JSJ2+C<4KGt!^QY1KhaEZyw+RWnR$
z5s%tKzNtZQXsrEK!@+ry4_03NU|I^3HIY^ol{c>F{SfWN1!rb<swx0rBV4uSrQ3bN
zb-I7H(hN=_nXcFsny0JWcy<_6GhetY^O5fyr3hXAum>Nq@{xv5QzHTMp8hHijsvyk
zrkFI^{5hEHp2D<1UIv_kyY{gPv*sCN`zM@kFx1ejQEqNx9OLkKGSNC)HmI}SSak6F
zT9D}z22Y|QecGqDEl%Ze!b-%ZU!^I%j|wJXJqwLL#Efz-^}(n1TD4i6vt%x&E-eFg
zQ}I}XSD={!GuT1;$5wq;mhRbo_eYsgl%s!<?NsIWjQt;Ch?Zce;{u2=W=*jKlS<J%
zb>Mli00x4tEt;p3$o;C&i(j?Z*^7fAESq>LJoEsJh{A9a&r&0zAtM9lI$hrz@hiB%
zp;qzD9nKMp{oPfldwH&W4m)WEit5oG1wlpnZ$I1V$?*j-w%6JSMJA?YUiivZz=v#N
zAEQmrHJs08o)Bz@0W2UK6ry-9SZ&toau)Y=pI^-<zb-KVk+_D{Hh8&;oAWfDqEI2a
zX-72RI(~S02lWl2%M|LnqK?K}L<p6NuHJb}$yGLKWj0mD1twQTE9ou0o3&nqR;jIP
zt&@r2NtO0cD(~Dc=2vTS3!Qib?W<7=1j)w{$r>Q{DaMn%^69VAN|(N`;x~CK)n$Cp
zzrUH`yqbHgj(vN7Yf$>|j^WKfAnIL6=Cz-2ABe&<_0I?|wK*y@6p<mcxnq3dE!zQl
zbC9Sjarzxno7;9ZuXV?9_qrw-XYc>2Dcw94wvJ%v3p&2EFqZWS)bR>a+?_n?=qbv&
z<*i@O4&tfH<;PA>YN&H@rg4QIdu!2H<-U%;Fgl(AQ+pIvRUEc5ZIsKE2x-J759PxV
zr_(x;1w~i3%2`E+#uc>%hwx8d*abjfkJEBp-{z0Ok}<|yKAT~5B;@q6Fkp~V!cMxP
zGfa&4H1>wY@<YHK-Abgu<K3w`AbNBA5Xg1z&jj(=jh=sEG4z=I=5HZG5Wpa|;sK_5
zsq3~&I{$~s=8=iXhPzDG4#=oR>e<FkVy0>|2dT~sWk~u9A>Lyf$QELQ#Zyn`dRx1A
z2<~}7e++a@d3)_D6cQ^9q8!_KPqzQ(NI<7qDrA+*6X~!Lr}TF1u+8CEMjPUUS%*9b
zq(){TYLBPNbkIyxnHR(NLx+TQgN7-%{Q13DY3qz^ac1WnNX~qTulO!<@uEq#TPF7Z
za2_V_^D^P=l`F)+_Y|%tAM37ylrCIY*FJJ&5!Y1SYU^I~DcIc6N|ejU(fY-dpU?3L
zhDNLMZE^qGXe+WeFl@ck;JV|bv^g{{%=NXe;l}4~1Fkng&yC0ioSe4wX2oezOZxbe
z=SsDRg6j|JYZNm~(}KRlI>(-+F6yG#P=A<8c1BYdr~g^A*MK1Q>k7`v(s8r|Cqvxs
z^mJIE5@`0C$l+)-Uf0&dxZ03lc{zK&TV0)X%}u8C@OUMwP{-YS;Kv)A#IVh{)s*{z
zXc9up483;fiAS|S(f#}x?RO;ua2aSl(GQX!u^r0V7Bt{_K!0T{u%NItTPwx=0jTxx
zcR4>V>?!hx+01jo1?*AF<0GHkr7h8&uE|7~R-sbfkec|HzAHU%N017uvBo?;t4slq
zL*px0lsHvL*h8+6a5F`R&5E-M3FeCp_5#WHi~d~7xZw@BuPg72@<dGXLyBL0*`jxu
zb^j-1k3m`xL{(~f_^YlM+8IQ)C?Vc%O2eOJqeZht7H+Rk?~eX-wPa`5DRyEmJmRW!
zl4MYe#XsUICeh3%5(kZooZMRlwh$@ak+IqwdayM84#Pp7I1(4#j_=qP=l;>_c&cZ*
zV(}eJS{>>YSytQ|^Z5x)lv<(w2n0F>A?Ep_%E}m$38?$!=E`a5*|ywZNx{~E5G7Y+
z=0@9pJsmif)eY3MapK~=5Ng;kFa|mc;*Z{U-?nMAW%k^3uV=E+v!j%;&IuAx&6p|j
zI|=ITleuf=EI3@UKdY`7%6lAGc~MHUS`5Ds^UgKCz45!_mwR!N*h@Egf~%ik_eA+w
zUyiMJ5Xz)z&;*ra2D4kz=Q2PHdI?p3l%8`JUijRPy@P%tm$`Gj1^pj<Gtd}E7LnLZ
z6Dn=z^~lP34Vk~GSPT}0sb1GngN`L@wSE#9JO>cwX_v$^rn6dYkG5g4O3%o?$4sv%
zQVIsSkHm|H8P<4iMx;H`i-TLF<&om}*XHBip*+N(q<L*{(s}hNGL|T?;)2CT6kL5)
zV^I}^hnBEx?Ikn}l!kl4#xI3<|H}Pp@Y6q!mvBm$yu92<)@9`?8RN&{3c2bW+h76z
z#qmZdxQY&aJcaYes?Jk=Xcc16)|R73j15|M%AuF70H2a^d)&#p)I~XLjBrH~Vbj&p
zAYydAZ<j$n?=$%y13~)rP^<sSjQ(F9(3Yyau(YjrUJJ?7PQXQ0r1=Kqukn`#Kk=tS
zLy#t||Ix4ge?$(x35WD)9~F()L~zk3uPgwk)6;hHV>OWjVWgBh%LvcB!yY@|PLRsv
zixC#SH9%h6IsRkDkRCPj4ApY<&`$#Uqb(CJNAwvlv@F(Z^Nr%4&A;rtg7d1y8d#@=
z{FtVh?hNw)7tLqT4eM=VoIqkR9#4I_yP8@)O>VMG>j0D<U|2b*8f18^dhBytR$%Yn
zb)f0tQDb30q6oh$Pna&uXT9Z|W9AuzcfN_4(p^{7|2!lP>G7FG^z_&H8j%F@+!f<I
zepdfT@?VCv={;*D&Y+|t2XdfE|A)fcp67?H-uVQ()QjxvP)syt85qY<c=-u`IaHO7
zo=Xt#+LUY{YBzEh_C#TBd(i2d^Nu`D!bu(Y%QLql9j7E~vD4Wv|0ooP87l&Njadu_
z+^}<1iKW=uTU~XaZGLY$i;@|QwljKYsev4OQ3g-VXvjOfwV$y_3H$xvB&{wlKb#{8
z+wbd9{jl6kNU`4D%$wCh|2KZ*0L-U7!lrQjxpJwm*C^>wHWT(NodO6od?6akXngT>
z{|WQLlVIhIdxH2x?vyL*)>06`D#2L;*AiYRha(B4J5CV0<XR-1`6GO%6zN+&Bapbj
z(&d5;susSJO$F@klaPF4O@1Ycnk|@o``K4b_Z#mf-CKe`aoJpOLN0LHaikJWdYiR0
zq%~`9i3@6y-3QOQli+u!znAgp2THV25+^4~--`R`R%eqqS4)#w<SUj~!@d(U;|D+U
z^A`e5OI^G=UdF#vKOVqrVm{h=9gkt^?9@A%o%NL06)7n&r61jHFcEEc_uU^;JI7n-
z^|4%JcujXo?r+^IHQ;RU<U{1%o{xvIX~~QH!y`=cPC!!3f!jDUP#SSC@}zqiQ~c4b
zgTe&C9U~I+Rq?kXz^!Gp)2IF++LNqdlJ1}?d&Km$xd46bvS626`qsu2Z$<;_-)mby
znw<11AKk_Jeit}UreAhA!)8(*)|%j$A-1kq68&oeoIv(&jBf2M>3&^}f{py&M*n@G
h)&B!Wf(5t%MLP?>dOGEC`qw|Ul;zds%3hg={vU3VU2^~c

literal 0
HcmV?d00001

diff --git a/keps/sig-node/4176-cpumanager-spread-cpus-preferred-policy/kep.yaml b/keps/sig-node/4176-cpumanager-spread-cpus-preferred-policy/kep.yaml
new file mode 100644
index 00000000000..cfe3d863033
--- /dev/null
+++ b/keps/sig-node/4176-cpumanager-spread-cpus-preferred-policy/kep.yaml
@@ -0,0 +1,42 @@
+title: New CPUManager Static Policy which spread hyperthreads across physical CPUs to better utilize CPU Cache
+kep-number: 4176
+authors:
+  - "@Jeffwan"
+  - "@LingyanYin"
+  - "@horacexd"
+  - "@LastNight1997"
+owning-sig: sig-node
+participating-sigs: []
+status: provisional
+creation-date: "2023-09-04"
+reviewers:
+  - TBD
+approvers:
+  - "@sig-node-tech-leads"
+see-also: []
+replaces: []
+
+# The target maturity stage in the current dev cycle for this KEP.
+stage: alpha
+
+# The most recent milestone for which work toward delivery of this KEP has been
+# done. This can be the current (upcoming) milestone, if it is being actively
+# worked on.
+latest-milestone: "v1.29"
+
+# The milestone at which this feature was, or is targeted to be, at each stage.
+milestone:
+  alpha: "v1.29"
+  beta: "v1.30"
+  stable: "v1.32"
+
+# The following PRR answers are required at alpha release
+# List the feature gate name and the components for which it must be enabled
+feature-gates:
+  - name: "CPUManagerPolicyAlphaOptions"
+    components:
+      - kubelet
+disable-supported: true
+
+# The following PRR answers are required at beta release
+metrics: []

From df0e0f7a17edc2d27d53e4010954d0c44ca97fa1 Mon Sep 17 00:00:00 2001
From: Jiaxin Shan <seedjeffwan@gmail.com>
Date: Tue, 12 Sep 2023 09:59:23 -0700
Subject: [PATCH 02/16] Address feedbacks from reviewers

---
 .../README.md                                             | 8 ++++++--
 .../4176-cpumanager-spread-cpus-preferred-policy/kep.yaml | 2 +-
 2 files changed, 7 insertions(+), 3 deletions(-)

diff --git a/keps/sig-node/4176-cpumanager-spread-cpus-preferred-policy/README.md b/keps/sig-node/4176-cpumanager-spread-cpus-preferred-policy/README.md
index eb7a955ac4d..ed7c00e33a3 100644
--- a/keps/sig-node/4176-cpumanager-spread-cpus-preferred-policy/README.md
+++ b/keps/sig-node/4176-cpumanager-spread-cpus-preferred-policy/README.md
@@ -174,7 +174,7 @@ updates.
 [documentation style guide]: https://github.com/kubernetes/community/blob/master/contributors/guide/style-guide.md
 -->
 
-We propose this KEP to introduce a new CPU Manager Static Policy that spreads hyper threads across physical cores. In this policy, we changed the cpu assignment sorting algorithm to sort by socket and then directly cpus without taking physical cores into ordering. It seems like a noisy neighbour issue, but it's always true. We will explain the reason in the motivation section. This policy is useful for some applications which need to take advantage of CPU Cache.
+We propose this KEP to introduce a new CPU Manager Static Policy Option that spreads hyper threads across physical cores. In this policy, we changed the cpu assignment sorting algorithm to sort by socket and then directly cpus without taking physical cores into ordering. It seems like a noisy neighbour issue, but it's always true. We will explain the reason in the motivation section. This policy is useful for some applications which need to take advantage of CPU Cache.
 
 ## Motivation
 
@@ -431,6 +431,11 @@ in back-to-back releases.
 - Deprecate the flag
 -->
 
+#### Alpha
+
+- Feature implemented behind the existing static policy feature flag
+- Initial e2e tests completed and enabled
+
 ### Upgrade / Downgrade Strategy
 
 <!--
@@ -567,7 +572,6 @@ https://github.com/kubernetes/kubernetes/pull/97058/files#diff-7826f7adbc1996a05
 <!--
 This section must be completed when targeting beta to a release.
 -->
-We only target alpha for this release.
 
 ###### How can a rollout or rollback fail? Can it impact already running workloads?
 
diff --git a/keps/sig-node/4176-cpumanager-spread-cpus-preferred-policy/kep.yaml b/keps/sig-node/4176-cpumanager-spread-cpus-preferred-policy/kep.yaml
index cfe3d863033..443988b9446 100644
--- a/keps/sig-node/4176-cpumanager-spread-cpus-preferred-policy/kep.yaml
+++ b/keps/sig-node/4176-cpumanager-spread-cpus-preferred-policy/kep.yaml
@@ -10,7 +10,7 @@ participating-sigs: []
 status: provisional
 creation-date: "2023-09-04"
 reviewers:
-  - TBD
+  - "@ffromani"
 approvers:
   - "@sig-node-tech-leads"
 see-also: []

From 981887390d3963ba60296d5c1612f22b78712394 Mon Sep 17 00:00:00 2001
From: Jiaxin Shan <seedjeffwan@gmail.com>
Date: Thu, 5 Oct 2023 01:46:26 -0700
Subject: [PATCH 03/16] Address reviewers 2nd review feedbacks

---
 .../README.md                                 | 106 ++++--------------
 1 file changed, 22 insertions(+), 84 deletions(-)

diff --git a/keps/sig-node/4176-cpumanager-spread-cpus-preferred-policy/README.md b/keps/sig-node/4176-cpumanager-spread-cpus-preferred-policy/README.md
index ed7c00e33a3..a67a49bebff 100644
--- a/keps/sig-node/4176-cpumanager-spread-cpus-preferred-policy/README.md
+++ b/keps/sig-node/4176-cpumanager-spread-cpus-preferred-policy/README.md
@@ -96,6 +96,7 @@ tags, and then generate with `hack/update-toc.sh`.
       - [Integration tests](#integration-tests)
       - [e2e tests](#e2e-tests)
   - [Graduation Criteria](#graduation-criteria)
+    - [Alpha](#alpha)
   - [Upgrade / Downgrade Strategy](#upgrade--downgrade-strategy)
   - [Version Skew Strategy](#version-skew-strategy)
 - [Production Readiness Review Questionnaire](#production-readiness-review-questionnaire)
@@ -332,7 +333,9 @@ This can inform certain test coverage improvements that we want to do before
 extending the production code to implement this enhancement.
 -->
 
-- `<package>`: `<date>` - `<test coverage>`
+- `k8s.io/kubernetes/pkg/kubelet/cm/cpumanager`: `20231005` - `86.3%`
+
+new added codes would be a cpu allocation policy option. We can follow how other options are tested and add enough unit tests.
 
 ##### Integration tests
 
@@ -351,7 +354,7 @@ For Beta and GA, add links to added tests together with links to k8s-triage for
 https://storage.googleapis.com/k8s-triage/index.html
 -->
 
-- <test>: <link to test coverage>
+No new integration tests for kubelet are planned.
 
 ##### e2e tests
 
@@ -365,76 +368,16 @@ https://storage.googleapis.com/k8s-triage/index.html
 We expect no non-infra related flakes in the last month as a GA graduation criteria.
 -->
 
-- <test>: <link to test coverage>
+No new e2e tests for kubelet are planned.
 
 ### Graduation Criteria
 
-<!--
-**Note:** *Not required until targeted at a release.*
-
-Define graduation milestones.
-
-These may be defined in terms of API maturity, [feature gate] graduations, or as
-something else. The KEP should keep this high-level with a focus on what
-signals will be looked at to determine graduation.
-
-Consider the following in developing the graduation criteria for this enhancement:
-- [Maturity levels (`alpha`, `beta`, `stable`)][maturity-levels]
-- [Feature gate][feature gate] lifecycle
-- [Deprecation policy][deprecation-policy]
-
-Clearly define what graduation means by either linking to the [API doc
-definition](https://kubernetes.io/docs/concepts/overview/kubernetes-api/#api-versioning)
-or by redefining what graduation means.
-
-In general we try to use the same stages (alpha, beta, GA), regardless of how the
-functionality is accessed.
-
-[feature gate]: https://git.k8s.io/community/contributors/devel/sig-architecture/feature-gates.md
-[maturity-levels]: https://git.k8s.io/community/contributors/devel/sig-architecture/api_changes.md#alpha-beta-and-stable-versions
-[deprecation-policy]: https://kubernetes.io/docs/reference/using-api/deprecation-policy/
-
-Below are some examples to consider, in addition to the aforementioned [maturity levels][maturity-levels].
-
-#### Alpha
-
-- Feature implemented behind a feature flag
-- Initial e2e tests completed and enabled
-
-#### Beta
-
-- Gather feedback from developers and surveys
-- Complete features A, B, C
-- Additional tests are in Testgrid and linked in KEP
-
-#### GA
-
-- N examples of real-world usage
-- N installs
-- More rigorous forms of testing—e.g., downgrade tests and scalability tests
-- Allowing time for feedback
-
-**Note:** Generally we also wait at least two releases between beta and
-GA/stable, because there's no opportunity for user feedback, or even bug reports,
-in back-to-back releases.
-
-**For non-optional features moving to GA, the graduation criteria must include
-[conformance tests].**
-
-[conformance tests]: https://git.k8s.io/community/contributors/devel/sig-architecture/conformance-tests.md
-
-#### Deprecation
-
-- Announce deprecation and support policy of the existing flag
-- Two versions passed since introducing the functionality that deprecates the flag (to address version skew)
-- Address feedback on usage/changed behavior, provided on GitHub issues
-- Deprecate the flag
--->
-
 #### Alpha
 
 - Feature implemented behind the existing static policy feature flag
-- Initial e2e tests completed and enabled
+- The functionality of new CPU allocation algorithm is implemented
+- Initial unit tests completed and coverage is improved
+- Documents is improved and enough guidance and examples can be given to potential users.
 
 ### Upgrade / Downgrade Strategy
 
@@ -647,8 +590,8 @@ Recall that end users cannot usually observe component logs or access metrics.
 - [ ] API .status
   - Condition name: 
   - Other field: 
-- [ ] Other (treat as last resort)
-  - Details:
+- [x] Other (treat as last resort)
+  - Details: Provide logical cpu allocation distribution across physical cores and also the cpu cache metrics from ecosystem.
 
 ###### What are the reasonable SLOs (Service Level Objectives) for the enhancement?
 
@@ -831,19 +774,10 @@ details). For now, we leave it here.
 N/A
 
 ###### What are other known failure modes?
-N/A
-<!--
-For each of them, fill in the following information by copying the below template:
-  - [Failure mode brief description]
-    - Detection: How can it be detected via metrics? Stated another way:
-      how can an operator troubleshoot without logging into a master or worker node?
-    - Mitigations: What can be done to stop the bleeding, especially for already
-      running user workloads?
-    - Diagnostics: What are the useful log messages and their required logging
-      levels that could help debug the issue?
-      Not required until feature graduated to beta.
-    - Testing: Are there any tests for failure mode? If not, describe why.
--->
+The failure modes is similar to existing options. It changes the way how cpu manager allocate CPUs.
+It's compatible when user switch between options, however, when the pod get rescheduled, it will follow the current static option instead of previous one. 
+
+When user switch to non static mode, then `/var/lib/kubelet/cpu_manager_state` requires deletion. This is a known compatibility issue.
 
 ###### What steps should be taken if SLOs are not being met to determine the problem?
 
@@ -862,9 +796,13 @@ Major milestones might include:
 
 ## Drawbacks
 
-<!--
-Why should this KEP _not_ be implemented?
--->
+Let's talk about the limitation of current policies.
+
+1. In a cluster with sparse workloads, we try to leverage as much cpu cache as we can. `full-pcpus-only` will always allocate full phsical cores and it introduces cache competition between vcpus.
+
+2. `distribute-cpus-across-num` will evenly distribut CPU across NUMA nodes. In some cases, we want the application to be allocated in single NUMA node if possible, which gives better performance.
+
+Existing solutions can not address all the special needs from high peformance applications, that's why a new option is needed.
 
 ## Alternatives
 

From 9072dfe74f1898d475cfa4c45adc119ca4e67d6e Mon Sep 17 00:00:00 2001
From: Jiaxin Shan <seedjeffwan@gmail.com>
Date: Thu, 5 Oct 2023 10:55:53 -0700
Subject: [PATCH 04/16] Address review feedback

---
 .../README.md                                 | 25 +++++++++++--------
 1 file changed, 14 insertions(+), 11 deletions(-)

diff --git a/keps/sig-node/4176-cpumanager-spread-cpus-preferred-policy/README.md b/keps/sig-node/4176-cpumanager-spread-cpus-preferred-policy/README.md
index a67a49bebff..7de14bc905e 100644
--- a/keps/sig-node/4176-cpumanager-spread-cpus-preferred-policy/README.md
+++ b/keps/sig-node/4176-cpumanager-spread-cpus-preferred-policy/README.md
@@ -335,8 +335,6 @@ extending the production code to implement this enhancement.
 
 - `k8s.io/kubernetes/pkg/kubelet/cm/cpumanager`: `20231005` - `86.3%`
 
-new added codes would be a cpu allocation policy option. We can follow how other options are tested and add enough unit tests.
-
 ##### Integration tests
 
 <!--
@@ -368,7 +366,15 @@ https://storage.googleapis.com/k8s-triage/index.html
 We expect no non-infra related flakes in the last month as a GA graduation criteria.
 -->
 
-No new e2e tests for kubelet are planned.
+These cases will be added in the existing `e2e_node` tests:
+  - CPU Manager works with `spread-physical-cpus-preferred` static policy option
+
+- Basic functionality
+1. Enable `CPUManagerPolicyAlphaOptions` and configure CPUManager policy option to `spread-physical-cpus-preferred`.
+2. Verify the machine has more than one physical cores.
+3. Create a simple pod with a container that requires 2 cpus.
+4. Verify that the container cpu allocation are across physical cores.
+6. Delete the pod.
 
 ### Graduation Criteria
 
@@ -591,7 +597,7 @@ Recall that end users cannot usually observe component logs or access metrics.
   - Condition name: 
   - Other field: 
 - [x] Other (treat as last resort)
-  - Details: Provide logical cpu allocation distribution across physical cores and also the cpu cache metrics from ecosystem.
+  - Details: Inspect the kubelet configuration of the nodes: check feature gate and usage of the new option.
 
 ###### What are the reasonable SLOs (Service Level Objectives) for the enhancement?
 
@@ -774,9 +780,12 @@ details). For now, we leave it here.
 N/A
 
 ###### What are other known failure modes?
+
 The failure modes is similar to existing options. It changes the way how cpu manager allocate CPUs.
 It's compatible when user switch between options, however, when the pod get rescheduled, it will follow the current static option instead of previous one. 
 
+Currently, in alpha version, we will think it's incompatile with other options. User should stick to this option. Compatibility issue would be resolved in future version.
+
 When user switch to non static mode, then `/var/lib/kubelet/cpu_manager_state` requires deletion. This is a known compatibility issue.
 
 ###### What steps should be taken if SLOs are not being met to determine the problem?
@@ -796,13 +805,7 @@ Major milestones might include:
 
 ## Drawbacks
 
-Let's talk about the limitation of current policies.
-
-1. In a cluster with sparse workloads, we try to leverage as much cpu cache as we can. `full-pcpus-only` will always allocate full phsical cores and it introduces cache competition between vcpus.
-
-2. `distribute-cpus-across-num` will evenly distribut CPU across NUMA nodes. In some cases, we want the application to be allocated in single NUMA node if possible, which gives better performance.
-
-Existing solutions can not address all the special needs from high peformance applications, that's why a new option is needed.
+This allocation strategy tries to avoid workload taking entire physical core and it is not suitable for all workloads. For example, if the workload is CPU intensive and it's not sensitive to CPU Cache, it's not suitable to use this policy. Otherwise, the application may suffer from performance regression.
 
 ## Alternatives
 

From 8d1542278103081666e975ba35bf6c8f5bf738ef Mon Sep 17 00:00:00 2001
From: Jiaxin Shan <seedjeffwan@gmail.com>
Date: Thu, 8 Feb 2024 23:28:21 +0800
Subject: [PATCH 05/16] Address reviewer comments for v1.30

---
 keps/prod-readiness/sig-node/4176.yaml                      | 6 ++++++
 .../4176-cpumanager-spread-cpus-preferred-policy/README.md  | 4 ++--
 .../4176-cpumanager-spread-cpus-preferred-policy/kep.yaml   | 6 +++---
 3 files changed, 11 insertions(+), 5 deletions(-)
 create mode 100644 keps/prod-readiness/sig-node/4176.yaml

diff --git a/keps/prod-readiness/sig-node/4176.yaml b/keps/prod-readiness/sig-node/4176.yaml
new file mode 100644
index 00000000000..29fe23b3fbb
--- /dev/null
+++ b/keps/prod-readiness/sig-node/4176.yaml
@@ -0,0 +1,6 @@
+# The KEP must have an approver from the
+# "prod-readiness-approvers" group
+# of http://git.k8s.io/enhancements/OWNERS_ALIASES
+kep-number: 4176
+alpha:
+  approver: "@jpbetz"
diff --git a/keps/sig-node/4176-cpumanager-spread-cpus-preferred-policy/README.md b/keps/sig-node/4176-cpumanager-spread-cpus-preferred-policy/README.md
index 7de14bc905e..d4972b23c27 100644
--- a/keps/sig-node/4176-cpumanager-spread-cpus-preferred-policy/README.md
+++ b/keps/sig-node/4176-cpumanager-spread-cpus-preferred-policy/README.md
@@ -501,7 +501,7 @@ If we reactivate the feature after a rollback, the outcome remains unchanged. Cu
 
 ###### Are there any tests for feature enablement/disablement?
 
-A dedicated e2e test will validate the preservation of the default behavior when the feature gate is turned off or when the feature is unused. This will be conducted through two distinct test scenarios.
+A dedicated e2e test will validate the preservation of the default behavior when the feature gate is turned off, when the feature is unused, when `pread-physical-cpus-preferred` turns off but `CPUManagerPolicyAlphaOptions` is disabled. This will be conducted through three distinct test scenarios.
 
 <!--
 The e2e framework does not currently support enabling or disabling feature
@@ -666,7 +666,7 @@ and creating new ones, as well as about cluster-level services (e.g. DNS):
       - Impact of its degraded performance or high-error rates on the feature:
 -->
 
-This applies to any machines with hyper-threading enabled.
+No. It doesn't rely on other Kubernetes components.
 
 ### Scalability
 
diff --git a/keps/sig-node/4176-cpumanager-spread-cpus-preferred-policy/kep.yaml b/keps/sig-node/4176-cpumanager-spread-cpus-preferred-policy/kep.yaml
index 443988b9446..34b72dfd9ca 100644
--- a/keps/sig-node/4176-cpumanager-spread-cpus-preferred-policy/kep.yaml
+++ b/keps/sig-node/4176-cpumanager-spread-cpus-preferred-policy/kep.yaml
@@ -26,9 +26,9 @@ latest-milestone: "v1.29"
 
 # The milestone at which this feature was, or is targeted to be, at each stage.
 milestone:
-  alpha: "v1.29"
-  beta: "v1.30"
-  stable: "v1.32"
+  alpha: "v1.30"
+  beta: "v1.31"
+  stable: "v1.33"
 
 # The following PRR answers are required at alpha release
 # List the feature gate name and the components for which it must be enabled

From 077567c8c808099f8d5ad211925991d5454c822b Mon Sep 17 00:00:00 2001
From: Jiaxin Shan <seedjeffwan@gmail.com>
Date: Thu, 8 Feb 2024 23:30:57 +0800
Subject: [PATCH 06/16] Update the lastest milestone to v1.30

---
 .../4176-cpumanager-spread-cpus-preferred-policy/kep.yaml       | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/keps/sig-node/4176-cpumanager-spread-cpus-preferred-policy/kep.yaml b/keps/sig-node/4176-cpumanager-spread-cpus-preferred-policy/kep.yaml
index 34b72dfd9ca..53a75b32cc9 100644
--- a/keps/sig-node/4176-cpumanager-spread-cpus-preferred-policy/kep.yaml
+++ b/keps/sig-node/4176-cpumanager-spread-cpus-preferred-policy/kep.yaml
@@ -22,7 +22,7 @@ stage: alpha
 # The most recent milestone for which work toward delivery of this KEP has been
 # done. This can be the current (upcoming) milestone, if it is being actively
 # worked on.
-latest-milestone: "v1.29"
+latest-milestone: "v1.30"
 
 # The milestone at which this feature was, or is targeted to be, at each stage.
 milestone:

From f3a655d21aeb9c81d41555c39b08792c5d726047 Mon Sep 17 00:00:00 2001
From: Jiaxin Shan <seedjeffwan@gmail.com>
Date: Fri, 9 Feb 2024 19:48:49 +0800
Subject: [PATCH 07/16] Update
 keps/sig-node/4176-cpumanager-spread-cpus-preferred-policy/README.md

Co-authored-by: Kevin Klues <klueska@gmail.com>
---
 .../4176-cpumanager-spread-cpus-preferred-policy/README.md      | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/keps/sig-node/4176-cpumanager-spread-cpus-preferred-policy/README.md b/keps/sig-node/4176-cpumanager-spread-cpus-preferred-policy/README.md
index d4972b23c27..1bb0cc66189 100644
--- a/keps/sig-node/4176-cpumanager-spread-cpus-preferred-policy/README.md
+++ b/keps/sig-node/4176-cpumanager-spread-cpus-preferred-policy/README.md
@@ -175,7 +175,7 @@ updates.
 [documentation style guide]: https://github.com/kubernetes/community/blob/master/contributors/guide/style-guide.md
 -->
 
-We propose this KEP to introduce a new CPU Manager Static Policy Option that spreads hyper threads across physical cores. In this policy, we changed the cpu assignment sorting algorithm to sort by socket and then directly cpus without taking physical cores into ordering. It seems like a noisy neighbour issue, but it's always true. We will explain the reason in the motivation section. This policy is useful for some applications which need to take advantage of CPU Cache.
+In this KEP, we propose a new CPU Manager Static Policy Option called `distribute-cores-across-cpus` to prefer allocating cores from different physical CPUs on the same socket. This new policy is analogous to the `distribute-cpus-across-numa` policy option in that it proposes to *spread* cores allocations out, rather than pack them together. The main difference being that this new policy spreads individual core allocations across CPUs, whereas the existing policy spreads them across NUMA nodes. Such a policy is useful, for example, if an application wants to avoid being a noisy neighbor with itself, but still take advantage of the L2 cache by running its application threads on the same socket.
 
 ## Motivation
 

From 564061da14d750aaedd3f59039442573826b9eec Mon Sep 17 00:00:00 2001
From: Jiaxin Shan <seedjeffwan@gmail.com>
Date: Fri, 9 Feb 2024 19:49:00 +0800
Subject: [PATCH 08/16] Update
 keps/sig-node/4176-cpumanager-spread-cpus-preferred-policy/README.md

Co-authored-by: Kevin Klues <klueska@gmail.com>
---
 .../4176-cpumanager-spread-cpus-preferred-policy/README.md      | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/keps/sig-node/4176-cpumanager-spread-cpus-preferred-policy/README.md b/keps/sig-node/4176-cpumanager-spread-cpus-preferred-policy/README.md
index 1bb0cc66189..bdc7b182db7 100644
--- a/keps/sig-node/4176-cpumanager-spread-cpus-preferred-policy/README.md
+++ b/keps/sig-node/4176-cpumanager-spread-cpus-preferred-policy/README.md
@@ -59,7 +59,7 @@ should be approved by the remaining approvers and/or the owning SIG (or
 SIG Architecture for cross-cutting KEPs).
 -->
 
-# KEP-4176: A new Static Policy to spread Hyperthreads across physical CPUs to better utilize CPU Cache
+# KEP-4176: A new static policy to prefer allocating cores from different CPUs on the same socket
 
 <!--
 This is the title of your KEP. Keep it short, simple, and descriptive. A good

From 39c49d206dc2806c4b56aba123a355de318b8052 Mon Sep 17 00:00:00 2001
From: Jiaxin Shan <seedjeffwan@gmail.com>
Date: Fri, 9 Feb 2024 23:33:41 +0800
Subject: [PATCH 09/16] Update
 keps/sig-node/4176-cpumanager-spread-cpus-preferred-policy/README.md

Co-authored-by: Kevin Klues <klueska@gmail.com>
---
 .../4176-cpumanager-spread-cpus-preferred-policy/README.md      | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/keps/sig-node/4176-cpumanager-spread-cpus-preferred-policy/README.md b/keps/sig-node/4176-cpumanager-spread-cpus-preferred-policy/README.md
index bdc7b182db7..8943fcd0e38 100644
--- a/keps/sig-node/4176-cpumanager-spread-cpus-preferred-policy/README.md
+++ b/keps/sig-node/4176-cpumanager-spread-cpus-preferred-policy/README.md
@@ -200,7 +200,7 @@ know that this has succeeded?
 - Introduce a new CPU Manager Static Policy that spreads hyper threads across physical cores without considering NUMA.
 - Enhance application performance by taking advantage of L2 Cache.
 
-### Non-Goals
+## Non-Goals
 
 <!--
 What is out of scope for this KEP? Listing non-goals helps to focus discussion

From f2294f4bf5c9fffdeb64ab582ed2e327f7ca5441 Mon Sep 17 00:00:00 2001
From: Jiaxin Shan <seedjeffwan@gmail.com>
Date: Fri, 9 Feb 2024 23:34:13 +0800
Subject: [PATCH 10/16] Update
 keps/sig-node/4176-cpumanager-spread-cpus-preferred-policy/README.md

Co-authored-by: Kevin Klues <klueska@gmail.com>
---
 .../4176-cpumanager-spread-cpus-preferred-policy/README.md      | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/keps/sig-node/4176-cpumanager-spread-cpus-preferred-policy/README.md b/keps/sig-node/4176-cpumanager-spread-cpus-preferred-policy/README.md
index 8943fcd0e38..a5caefeeaf3 100644
--- a/keps/sig-node/4176-cpumanager-spread-cpus-preferred-policy/README.md
+++ b/keps/sig-node/4176-cpumanager-spread-cpus-preferred-policy/README.md
@@ -197,7 +197,7 @@ demonstrate the interest in a KEP within the wider Kubernetes community.
 List the specific goals of the KEP. What is it trying to achieve? How will we
 know that this has succeeded?
 -->
-- Introduce a new CPU Manager Static Policy that spreads hyper threads across physical cores without considering NUMA.
+- Introduce a new CPU Manager Static Policy that spreads CPUs across physical cores without considering NUMA.
 - Enhance application performance by taking advantage of L2 Cache.
 
 ## Non-Goals

From 8aff41fad155270c215d252f1a8041620bcdede8 Mon Sep 17 00:00:00 2001
From: Jiaxin Shan <seedjeffwan@gmail.com>
Date: Fri, 9 Feb 2024 23:34:53 +0800
Subject: [PATCH 11/16] Update
 keps/sig-node/4176-cpumanager-spread-cpus-preferred-policy/README.md

Co-authored-by: Kevin Klues <klueska@gmail.com>
---
 .../4176-cpumanager-spread-cpus-preferred-policy/README.md      | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/keps/sig-node/4176-cpumanager-spread-cpus-preferred-policy/README.md b/keps/sig-node/4176-cpumanager-spread-cpus-preferred-policy/README.md
index a5caefeeaf3..c942922487b 100644
--- a/keps/sig-node/4176-cpumanager-spread-cpus-preferred-policy/README.md
+++ b/keps/sig-node/4176-cpumanager-spread-cpus-preferred-policy/README.md
@@ -207,7 +207,7 @@ What is out of scope for this KEP? Listing non-goals helps to focus discussion
 and make progress.
 -->
 
-- This proposal does not aim to modify the existing CPU Manager Core Binding Policies. It focuses solely on introducing a new policy for spreading hyper threads across physical cores.
+- This proposal does not aim to modify the existing CPU Manager Core Binding Policies. It focuses solely on introducing a new policy for spreading CPUs across physical cores.
 - It does not address other resource allocation or management aspects within Kubernetes.
 
 

From b44a5fdcb44d054dacd70533f85f3d496bbf8f3e Mon Sep 17 00:00:00 2001
From: Jiaxin Shan <seedjeffwan@gmail.com>
Date: Fri, 9 Feb 2024 23:35:08 +0800
Subject: [PATCH 12/16] Update
 keps/sig-node/4176-cpumanager-spread-cpus-preferred-policy/README.md

Co-authored-by: Kevin Klues <klueska@gmail.com>
---
 .../README.md                                    | 16 +---------------
 1 file changed, 1 insertion(+), 15 deletions(-)

diff --git a/keps/sig-node/4176-cpumanager-spread-cpus-preferred-policy/README.md b/keps/sig-node/4176-cpumanager-spread-cpus-preferred-policy/README.md
index c942922487b..13fd40a2265 100644
--- a/keps/sig-node/4176-cpumanager-spread-cpus-preferred-policy/README.md
+++ b/keps/sig-node/4176-cpumanager-spread-cpus-preferred-policy/README.md
@@ -177,21 +177,7 @@ updates.
 
 In this KEP, we propose a new CPU Manager Static Policy Option called `distribute-cores-across-cpus` to prefer allocating cores from different physical CPUs on the same socket. This new policy is analogous to the `distribute-cpus-across-numa` policy option in that it proposes to *spread* cores allocations out, rather than pack them together. The main difference being that this new policy spreads individual core allocations across CPUs, whereas the existing policy spreads them across NUMA nodes. Such a policy is useful, for example, if an application wants to avoid being a noisy neighbor with itself, but still take advantage of the L2 cache by running its application threads on the same socket.
 
-## Motivation
-
-<!--
-This section is for explicitly listing the motivation, goals, and non-goals of
-this KEP.  Describe why the change is important and the benefits to users. The
-motivation section can optionally provide links to [experience reports] to
-demonstrate the interest in a KEP within the wider Kubernetes community.
-
-[experience reports]: https://github.com/golang/go/wiki/ExperienceReports
--->
-
-`full-pcpus-only` was introduced to resolve the problem that different containers sharing the same physical cores which leads noisy neighbor issue. `distribute-cpus-across-numa` distributed cpus across NUMA to make sure no single worker suffers from NUMA effects more than any other, improving the overall performance of these types of applications. These two and default behavior can not meet the requirement of some applications which need to take advantage of L2 Cache. In that case, we need to spread hyper threads across physical cores, while NUMA is not a important factor. In such cases, the assumption is that the single physical core with two applications won't be always busy. So we can take advantage of CPU Cache to improve the performance of these applications. Otherwise, it still have the noisy neighbor issue.
-
-
-### Goals
+## Goals
 
 <!--
 List the specific goals of the KEP. What is it trying to achieve? How will we

From f3c867ae163ecaddf0ddc7051e434039f9d185a8 Mon Sep 17 00:00:00 2001
From: Jiaxin Shan <seedjeffwan@gmail.com>
Date: Fri, 9 Feb 2024 23:35:19 +0800
Subject: [PATCH 13/16] Update
 keps/sig-node/4176-cpumanager-spread-cpus-preferred-policy/README.md

Co-authored-by: Kevin Klues <klueska@gmail.com>
---
 .../4176-cpumanager-spread-cpus-preferred-policy/README.md      | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/keps/sig-node/4176-cpumanager-spread-cpus-preferred-policy/README.md b/keps/sig-node/4176-cpumanager-spread-cpus-preferred-policy/README.md
index 13fd40a2265..fbdc2273840 100644
--- a/keps/sig-node/4176-cpumanager-spread-cpus-preferred-policy/README.md
+++ b/keps/sig-node/4176-cpumanager-spread-cpus-preferred-policy/README.md
@@ -175,7 +175,7 @@ updates.
 [documentation style guide]: https://github.com/kubernetes/community/blob/master/contributors/guide/style-guide.md
 -->
 
-In this KEP, we propose a new CPU Manager Static Policy Option called `distribute-cores-across-cpus` to prefer allocating cores from different physical CPUs on the same socket. This new policy is analogous to the `distribute-cpus-across-numa` policy option in that it proposes to *spread* cores allocations out, rather than pack them together. The main difference being that this new policy spreads individual core allocations across CPUs, whereas the existing policy spreads them across NUMA nodes. Such a policy is useful, for example, if an application wants to avoid being a noisy neighbor with itself, but still take advantage of the L2 cache by running its application threads on the same socket.
+In this KEP, we propose a new CPU Manager Static Policy Option called `distribute-cpus-across-cores` to prefer allocating CPUs from different physical cores on the same socket. This new policy is analogous to the `distribute-cpus-across-numa` policy option in that it proposes to *spread* CPU allocations out, rather than pack them together. The main difference being that this new policy spreads individual CPU allocations across cores, whereas the existing policy spreads them across NUMA nodes. Such a policy is useful, for example, if an application wants to avoid being a noisy neighbor with itself, but still take advantage of the L2 cache by running its threads on the same socket.
 
 ## Goals
 

From 4d7f4f501925875ba390c8a47c7321765ff77f8d Mon Sep 17 00:00:00 2001
From: Jiaxin Shan <seedjeffwan@gmail.com>
Date: Fri, 9 Feb 2024 23:35:29 +0800
Subject: [PATCH 14/16] Update
 keps/sig-node/4176-cpumanager-spread-cpus-preferred-policy/README.md

Co-authored-by: Kevin Klues <klueska@gmail.com>
---
 .../4176-cpumanager-spread-cpus-preferred-policy/README.md      | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/keps/sig-node/4176-cpumanager-spread-cpus-preferred-policy/README.md b/keps/sig-node/4176-cpumanager-spread-cpus-preferred-policy/README.md
index fbdc2273840..17e0a361758 100644
--- a/keps/sig-node/4176-cpumanager-spread-cpus-preferred-policy/README.md
+++ b/keps/sig-node/4176-cpumanager-spread-cpus-preferred-policy/README.md
@@ -221,7 +221,7 @@ bogged down.
 
 #### Story 1 Bytedance Database Performance Optimization
 
-We're running DB instances in Kubernetes and adopt default static policy in the past. While, we notice that the performance of DB instances is not stable. If an instance is under pressure, in original way, it was allocated two cpus from same physical core. However, an important pattern we notice is not always all instances are busy. After exploration, we find that the CPU cache is one bottleneck, once we spread hyper threads across physical cores, the busy instance can leverage more CPU cache and performance is improved a lot.
+We're running DB instances in Kubernetes and adopt default static policy in the past. While, we notice that the performance of DB instances is not stable. If an instance is under pressure, in original way, it was allocated two CPUs from same physical core. However, an important pattern we notice is not always all instances are busy. After exploration, we find that the CPU cache is one bottleneck, once we allocate CPUs across physical cores, the busy instance can leverage more CPU cache and performance is improved a lot.
 
 #### Story 2
 

From 41739e90d60df52acd1f35ff73689485a359bcad Mon Sep 17 00:00:00 2001
From: Jiaxin Shan <seedjeffwan@gmail.com>
Date: Fri, 9 Feb 2024 23:35:50 +0800
Subject: [PATCH 15/16] Update
 keps/sig-node/4176-cpumanager-spread-cpus-preferred-policy/README.md

Co-authored-by: Kevin Klues <klueska@gmail.com>
---
 .../4176-cpumanager-spread-cpus-preferred-policy/README.md      | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/keps/sig-node/4176-cpumanager-spread-cpus-preferred-policy/README.md b/keps/sig-node/4176-cpumanager-spread-cpus-preferred-policy/README.md
index 17e0a361758..bef1e6fe2d7 100644
--- a/keps/sig-node/4176-cpumanager-spread-cpus-preferred-policy/README.md
+++ b/keps/sig-node/4176-cpumanager-spread-cpus-preferred-policy/README.md
@@ -208,7 +208,7 @@ The "Design Details" section below is for the real
 nitty-gritty.
 -->
 
-We propose to add a new `CPUManager` policy option called `spread-physical-cpus-preferred` to the static CPUManager policy. When enabled, this will trigger the CPUManager to try to allocate CPUs across physical nodes as much as possible.
+We propose to add a new `CPUManager` policy option called `distribute-cpus-across-cores` to the static CPUManager policy. When enabled, this will trigger the CPUManager to try to allocate CPUs across physical cores as much as possible. It will not prohibit a CPU from being allocated on a core that already has a CPU allocated, but it will only resort to doing so once there is no other option.
 
 ### User Stories (Optional)
 

From 5a987826acefde8e52ff3ea6e97ca7de4eaedf98 Mon Sep 17 00:00:00 2001
From: Jiaxin Shan <seedjeffwan@gmail.com>
Date: Fri, 9 Feb 2024 23:44:04 +0800
Subject: [PATCH 16/16] Update kep status and toc

---
 .../4176-cpumanager-spread-cpus-preferred-policy/README.md   | 5 ++---
 .../4176-cpumanager-spread-cpus-preferred-policy/kep.yaml    | 2 +-
 2 files changed, 3 insertions(+), 4 deletions(-)

diff --git a/keps/sig-node/4176-cpumanager-spread-cpus-preferred-policy/README.md b/keps/sig-node/4176-cpumanager-spread-cpus-preferred-policy/README.md
index bef1e6fe2d7..fb115af290c 100644
--- a/keps/sig-node/4176-cpumanager-spread-cpus-preferred-policy/README.md
+++ b/keps/sig-node/4176-cpumanager-spread-cpus-preferred-policy/README.md
@@ -80,9 +80,8 @@ tags, and then generate with `hack/update-toc.sh`.
 <!-- toc -->
 - [Release Signoff Checklist](#release-signoff-checklist)
 - [Summary](#summary)
-- [Motivation](#motivation)
-  - [Goals](#goals)
-  - [Non-Goals](#non-goals)
+- [Goals](#goals)
+- [Non-Goals](#non-goals)
 - [Proposal](#proposal)
   - [User Stories (Optional)](#user-stories-optional)
     - [Story 1 Bytedance Database Performance Optimization](#story-1-bytedance-database-performance-optimization)
diff --git a/keps/sig-node/4176-cpumanager-spread-cpus-preferred-policy/kep.yaml b/keps/sig-node/4176-cpumanager-spread-cpus-preferred-policy/kep.yaml
index 53a75b32cc9..15a70b70745 100644
--- a/keps/sig-node/4176-cpumanager-spread-cpus-preferred-policy/kep.yaml
+++ b/keps/sig-node/4176-cpumanager-spread-cpus-preferred-policy/kep.yaml
@@ -7,7 +7,7 @@ authors:
   - "@LastNight1997"
 owning-sig: sig-node
 participating-sigs: []
-status: provisional
+status: implementable
 creation-date: "2023-09-04"
 reviewers:
   - "@ffromani"