You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Add support for CPU and memory affinity on windows by enabling the cpu, memory and topology managers for Windows, which are currently not enabled.
195
192
196
-
Enables Low latency workloads co-hosted on the same nodes in Windows Server show noisy neighbor behavior preventing them to achieve their performance goals. This feature is needed to add the necessary isolation to accomplish both high performance and co-hosting efficiency. The feature is enabled and available in Linux and Windows users are asking for the same features on Windows.
193
+
Add support for CPU and memory affinity for Windows nodes by enabling the cpu, memory and topology managers for Windows,
194
+
which are currently not enabled.
195
+
196
+
Enabling Low latency workloads co-hosted on the same nodes in Windows Server show noisy neighbor behaviors
197
+
preventing them from achieving their expected performance goals.
198
+
This feature is needed to add the necessary isolation to accomplish both high performance and co-hosting efficiency.
199
+
The feature is enabled and available in Linux and Windows users are asking for the same features on Windows.
197
200
198
201
### Goals
199
202
200
203
<!--
201
204
List the specific goals of the KEP. What is it trying to achieve? How will we
202
205
know that this has succeeded?
203
206
-->
204
-
- Enable CPU manager for Windows allowing for CPU affinity
205
-
- Enable Memory Manager for Windows allowing for Memory Affinity
206
-
- Enable Topology Manager for Windows allowing for coordination of Memory and CPU affinity
207
+
- Enable CPU manager for Windows allowing for CPU affinity for configured pods
208
+
- Enable Memory Manager for Windows allowing for memory affinity for configured pods
209
+
- Enable Topology Manager for Windows allowing for coordination of Memory and CPU affinity at the node level for scheduled pods
207
210
208
211
### Non-Goals
209
212
@@ -229,7 +232,7 @@ nitty-gritty.
229
232
The proposal requires very little changes to the code for the managers and instead extends the [Windows](https://learn.microsoft.com/en-us/windows/win32/procthread/processor-groups) concepts to a CAdvisor mapping to enable the [topology structure in kubelet](https://github.com/kubernetes/kubernetes/blob/cede96336a809a67546ca08df0748e4253ec270d/pkg/kubelet/cm/cpumanager/topology/topology.go#L34-L39).
230
233
231
234
There are no plans to change the core logic for selecting CPU's and NUMA nodes in the CPU/Memory/Tolopology managers from the existing KEPS ([memory-manager](keps/sig-node/1769-memory-manager)/[cpu-manager](keps/sig-node/3570-cpu-manager)/[topology-manager](keps/sig-node/693-topology-manager")). The logic is currently in platform agnostic
232
-
structure so the selection process is does not require changes to adapt for Windows. The Windows specific considerations for each of the managers will be covered in separate sections in this document.
235
+
structures so the selection process is does not require changes for adoption on Windows. The Windows specific considerations for each of the managers will be covered in separate sections in this document.
233
236
234
237
235
238
### User Stories (Optional)
@@ -241,9 +244,11 @@ the system. The goal here is to make this feel real for users without getting
0 commit comments