github/kata-containers
1
0
Fork 2
mirror of https://github.com/kata-containers/kata-containers.git synced 2025-09-01 17:06:28 +00:00
Code Issues Projects Releases Wiki Activity

gpu: Update runtimeClasses for correct podoverhead

Browse Source 
We cannot only rely only on default_cpu and default_memory in the
config, default is 1 and 2Gi but we need some overhead for QEMU and
the other related binaries running as the pod overhead. Especially
when QEMU is hot-plugging GPUs, CPUs, and memory it can consume more
memory.

Signed-off-by: Zvonko Kaiser <zkaiser@nvidia.com>
This commit is contained in:
Zvonko Kaiser
2025-05-01 15:11:42 +00:00
parent 06466a53c5
commit 9f5e700a93
3 changed files with 6 additions and 6 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
tools/packaging/kata-deploy/runtimeclasses/kata-qemu-nvidia-gpu-snp.yaml
View File

@@ -6,8 +6,8 @@ metadata:
handler: kata-qemu-nvidia-gpu-snp handler: kata-qemu-nvidia-gpu-snp
overhead: overhead:
podFixed: podFixed:
memory: "2048Mi" memory: "4096Mi"
cpu: "1.0" cpu: "1"
scheduling: scheduling:
nodeSelector: nodeSelector:
katacontainers.io/kata-runtime: "true" katacontainers.io/kata-runtime: "true"

4
tools/packaging/kata-deploy/runtimeclasses/kata-qemu-nvidia-gpu-tdx.yaml
View File

@@ -6,8 +6,8 @@ metadata:
handler: kata-qemu-nvidia-gpu-tdx handler: kata-qemu-nvidia-gpu-tdx
overhead: overhead:
podFixed: podFixed:
memory: "2048Mi" memory: "4096Mi"
cpu: "1.0" cpu: "1"
scheduling: scheduling:
nodeSelector: nodeSelector:
katacontainers.io/kata-runtime: "true" katacontainers.io/kata-runtime: "true"

4
tools/packaging/kata-deploy/runtimeclasses/kata-qemu-nvidia-gpu.yaml
View File

@@ -6,8 +6,8 @@ metadata:
handler: kata-qemu-nvidia-gpu handler: kata-qemu-nvidia-gpu
overhead: overhead:
podFixed: podFixed:
memory: "160Mi" memory: "4096Mi"
cpu: "250m" cpu: "1"
scheduling: scheduling:
nodeSelector: nodeSelector:
katacontainers.io/kata-runtime: "true" katacontainers.io/kata-runtime: "true"