Kubernetes叢集部署Node Feature Discovery元件用於檢測叢集節點特性

人艰不拆_zmc發表於2024-03-14

1、概述

Node Feature Discovery(NFD)是由Intel建立的專案,能夠幫助Kubernetes叢集更智慧地管理節點資源。它透過檢測每個節點的特效能力(例如CPU型號、GPU型號、記憶體大小等)並將這些能力以標籤的形式傳送到Kubernetes叢集的API伺服器(kube-apiserver)。然後,透過kube-apiserver修改節點的標籤。這些標籤可以幫助排程器(kube-scheduler)更智慧地選擇最適合特定工作負載的節點來執行Pod。

Github:https://github.com/kubernetes-sigs/node-feature-discovery
Docs:https://kubernetes-sigs.github.io/node-feature-discovery/master/get-started/index.html

2、元件架構

NFD 細分為 NFD-Master 和 NFD-Worker 兩個元件:

NFD-Master:是一個負責與 kubernetes API Server 通訊的Deployment Pod,它從 NFD-Worker 接收節點特性並相應地修改 Node 資源物件(標籤、註解)。

NFD-Worker:是一個負責對 Node 的特效能力進行檢測的 Daemon Pod,然後它將資訊傳遞給 NFD-Master,NFD-Worker 應該在每個 Node 上執行。

可以檢測發現的硬體特徵源(feature sources)清單包括:

  • CPU
  • IOMMU
  • Kernel
  • Memory
  • Network
  • PCI
  • Storage
  • System
  • USB
  • Custom (rule-based custom features)
  • Local (hooks for user-specific features)

3、元件安裝

(1)安裝前檢視叢集節點狀態

[root@master-10 ~]# kubectl get nodes
NAME                  STATUS   ROLES                         AGE   VERSION
master-10.20.31.105   Ready    control-plane,master,worker   31h   v1.21.5

節點詳細資訊,主要關注標籤、註解。

[root@master-10 ~]# kubectl describe nodes master-10.20.31.105 
Name:               master-10.20.31.105
Roles:              control-plane,master,worker
Labels:             beta.kubernetes.io/arch=amd64
                    beta.kubernetes.io/os=linux
                    kubernetes.io/arch=amd64
                    kubernetes.io/hostname=master-10.20.31.105
                    kubernetes.io/os=linux
                    node-role.kubernetes.io/control-plane=
                    node-role.kubernetes.io/master=
                    node-role.kubernetes.io/worker=
                    node.kubernetes.io/exclude-from-external-load-balancers=
Annotations:        flannel.alpha.coreos.com/backend-data: {"VtepMAC":"c6:fb:4b:8a:bb:12"}
                    flannel.alpha.coreos.com/backend-type: vxlan
                    flannel.alpha.coreos.com/kube-subnet-manager: true
                    flannel.alpha.coreos.com/public-ip: 10.20.31.105
                    kubeadm.alpha.kubernetes.io/cri-socket: /var/run/dockershim.sock
                    node.alpha.kubernetes.io/ttl: 0
                    volumes.kubernetes.io/controller-managed-attach-detach: true
CreationTimestamp:  Tue, 12 Mar 2024 21:01:31 -0400
Taints:             <none>
........

(2)元件安裝

[root@master-10 opt]# kubectl apply -k https://github.com/kubernetes-sigs/node-feature-discovery/deployment/overlays/default?ref=v0.14.2
namespace/node-feature-discovery created
customresourcedefinition.apiextensions.k8s.io/nodefeaturerules.nfd.k8s-sigs.io created
customresourcedefinition.apiextensions.k8s.io/nodefeatures.nfd.k8s-sigs.io created
serviceaccount/nfd-master created
serviceaccount/nfd-worker created
role.rbac.authorization.k8s.io/nfd-worker created
clusterrole.rbac.authorization.k8s.io/nfd-master created
rolebinding.rbac.authorization.k8s.io/nfd-worker created
clusterrolebinding.rbac.authorization.k8s.io/nfd-master created
configmap/nfd-master-conf created
configmap/nfd-worker-conf created
service/nfd-master created
deployment.apps/nfd-master created
daemonset.apps/nfd-worker created

(3)檢視元件狀態

[root@master-10 opt]# kubectl get pods -n=node-feature-discovery 
NAME                          READY   STATUS    RESTARTS   AGE
nfd-master-5c4684f5cb-hvjjb   1/1     Running   0          4m11s
nfd-worker-cpwx6              1/1     Running   0          4m11s

(4)檢視元件日誌

可以看到nfd-worker元件預設每隔一分鐘檢測一次節點特性。

[root@master-10 ~]# kubectl logs -f -n=node-feature-discovery nfd-worker-rlf5t 
I0314 06:30:32.003264       1 main.go:66] "-server is deprecated, will be removed in a future release along with the deprecated gRPC API"
I0314 06:30:32.003372       1 nfd-worker.go:219] "Node Feature Discovery Worker" version="v0.14.2" nodeName="master-10.20.31.105" namespace="node-feature-discovery"
I0314 06:30:32.003589       1 nfd-worker.go:520] "configuration file parsed" path="/etc/kubernetes/node-feature-discovery/nfd-worker.conf"
I0314 06:30:32.004500       1 nfd-worker.go:552] "configuration successfully updated" configuration={"Core":{"Klog":{},"LabelWhiteList":{},"NoPublish":false,"FeatureSources":["all"],"Sources":null,"LabelSources":["all"],"SleepInterval":{"Duration":60000000000}},"Sources":{"cpu":{"cpuid":{"attributeBlacklist":["BMI1","BMI2","CLMUL","CMOV","CX16","ERMS","F16C","HTT","LZCNT","MMX","MMXEXT","NX","POPCNT","RDRAND","RDSEED","RDTSCP","SGX","SGXLC","SSE","SSE2","SSE3","SSE4","SSE42","SSSE3","TDX_GUEST"]}},"custom":[],"fake":{"labels":{"fakefeature1":"true","fakefeature2":"true","fakefeature3":"true"},"flagFeatures":["flag_1","flag_2","flag_3"],"attributeFeatures":{"attr_1":"true","attr_2":"false","attr_3":"10"},"instanceFeatures":[{"attr_1":"true","attr_2":"false","attr_3":"10","attr_4":"foobar","name":"instance_1"},{"attr_1":"true","attr_2":"true","attr_3":"100","name":"instance_2"},{"name":"instance_3"}]},"kernel":{"KconfigFile":"","configOpts":["NO_HZ","NO_HZ_IDLE","NO_HZ_FULL","PREEMPT"]},"local":{},"pci":{"deviceClassWhitelist":["03","0b40","12"],"deviceLabelFields":["class","vendor"]},"usb":{"deviceClassWhitelist":["0e","ef","fe","ff"],"deviceLabelFields":["class","vendor","device"]}}}
I0314 06:30:32.004796       1 metrics.go:70] "metrics server starting" port=8081
I0314 06:30:32.019135       1 nfd-worker.go:562] "starting feature discovery..."
I0314 06:30:32.019364       1 nfd-worker.go:577] "feature discovery completed"
I0314 06:31:32.021520       1 nfd-worker.go:562] "starting feature discovery..."
I0314 06:31:32.021695       1 nfd-worker.go:577] "feature discovery completed"
I0314 06:32:32.027970       1 nfd-worker.go:562] "starting feature discovery..."
I0314 06:32:32.028141       1 nfd-worker.go:577] "feature discovery completed"

可以看到nfd-master元件啟動後預設第一分鐘相應地修改 Node 資源物件(標籤、註解),之後是每隔一個小時修改一次 Node 資源物件(標籤、註解),也就是說如果一個小時以內使用者手動誤修改node資源特性資訊(標籤、註解),最多需要一個小時nfd-master元件才自動更正node資源特性資訊。

[root@master-10 ~]# kubectl logs -n=node-feature-discovery nfd-master-5c4684f5cb-hvjjb 
I0314 06:23:08.190218       1 nfd-master.go:213] "Node Feature Discovery Master" version="v0.14.2" nodeName="master-10.20.31.105" namespace="node-feature-discovery"
I0314 06:23:08.190356       1 nfd-master.go:1214] "configuration file parsed" path="/etc/kubernetes/node-feature-discovery/nfd-master.conf"
I0314 06:23:08.190912       1 nfd-master.go:1274] "configuration successfully updated" configuration=<
	DenyLabelNs: {}
	EnableTaints: false
	ExtraLabelNs: {}
	Klog: {}
	LabelWhiteList: {}
	LeaderElection:
	  LeaseDuration:
	    Duration: 15000000000
	  RenewDeadline:
	    Duration: 10000000000
	  RetryPeriod:
	    Duration: 2000000000
	NfdApiParallelism: 10
	NoPublish: false
	ResourceLabels: {}
	ResyncPeriod:
	  Duration: 3600000000000
 >
I0314 06:23:08.190928       1 nfd-master.go:1338] "starting the nfd api controller"
I0314 06:23:08.191105       1 node-updater-pool.go:79] "starting the NFD master node updater pool" parallelism=10
I0314 06:23:08.860810       1 metrics.go:115] "metrics server starting" port=8081
I0314 06:23:08.861033       1 component.go:36] [core][Server #1] Server created
I0314 06:23:08.861050       1 nfd-master.go:347] "gRPC server serving" port=8080
I0314 06:23:08.861084       1 component.go:36] [core][Server #1 ListenSocket #2] ListenSocket created
I0314 06:23:09.860886       1 nfd-master.go:694] "will process all nodes in the cluster"
I0314 06:23:09.923362       1 nfd-master.go:1086] "node updated" nodeName="master-10.20.31.105"
I0314 07:23:09.224254       1 nfd-master.go:1086] "node updated" nodeName="master-10.20.31.105"
I0314 08:23:09.081362       1 nfd-master.go:1086] "node updated" nodeName="master-10.20.31.105"

(5)檢視節點特性資訊

可以看到NFD元件已經把節點特性資訊維護到了節點標籤、註解上,其中標籤字首預設為 feature.node.kubernetes.io/。

[root@master-10 opt]# kubectl describe node master-10.20.31.105 
Name:               master-10.20.31.105
Roles:              control-plane,master,worker
Labels:             beta.kubernetes.io/arch=amd64
                    beta.kubernetes.io/os=linux
                    feature.node.kubernetes.io/cpu-cpuid.ADX=true
                    feature.node.kubernetes.io/cpu-cpuid.AESNI=true
                    feature.node.kubernetes.io/cpu-cpuid.AVX=true
                    feature.node.kubernetes.io/cpu-cpuid.AVX2=true
                    feature.node.kubernetes.io/cpu-cpuid.AVX512BW=true
                    feature.node.kubernetes.io/cpu-cpuid.AVX512CD=true
                    feature.node.kubernetes.io/cpu-cpuid.AVX512DQ=true
                    feature.node.kubernetes.io/cpu-cpuid.AVX512F=true
                    feature.node.kubernetes.io/cpu-cpuid.AVX512VL=true
                    feature.node.kubernetes.io/cpu-cpuid.CMPXCHG8=true
                    feature.node.kubernetes.io/cpu-cpuid.FMA3=true
                    feature.node.kubernetes.io/cpu-cpuid.FXSR=true
                    feature.node.kubernetes.io/cpu-cpuid.FXSROPT=true
                    feature.node.kubernetes.io/cpu-cpuid.HLE=true
                    feature.node.kubernetes.io/cpu-cpuid.HYPERVISOR=true
                    feature.node.kubernetes.io/cpu-cpuid.LAHF=true
                    feature.node.kubernetes.io/cpu-cpuid.MOVBE=true
                    feature.node.kubernetes.io/cpu-cpuid.MPX=true
                    feature.node.kubernetes.io/cpu-cpuid.OSXSAVE=true
                    feature.node.kubernetes.io/cpu-cpuid.RTM=true
                    feature.node.kubernetes.io/cpu-cpuid.SYSCALL=true
                    feature.node.kubernetes.io/cpu-cpuid.SYSEE=true
                    feature.node.kubernetes.io/cpu-cpuid.X87=true
                    feature.node.kubernetes.io/cpu-cpuid.XSAVE=true
                    feature.node.kubernetes.io/cpu-cpuid.XSAVEC=true
                    feature.node.kubernetes.io/cpu-cpuid.XSAVEOPT=true
                    feature.node.kubernetes.io/cpu-cpuid.XSAVES=true
                    feature.node.kubernetes.io/cpu-hardware_multithreading=false
                    feature.node.kubernetes.io/cpu-model.family=6
                    feature.node.kubernetes.io/cpu-model.id=85
                    feature.node.kubernetes.io/cpu-model.vendor_id=Intel
                    feature.node.kubernetes.io/kernel-config.NO_HZ=true
                    feature.node.kubernetes.io/kernel-config.NO_HZ_FULL=true
                    feature.node.kubernetes.io/kernel-version.full=3.10.0-1160.105.1.el7.x86_64
                    feature.node.kubernetes.io/kernel-version.major=3
                    feature.node.kubernetes.io/kernel-version.minor=10
                    feature.node.kubernetes.io/kernel-version.revision=0
                    feature.node.kubernetes.io/pci-0300_15ad.present=true
                    feature.node.kubernetes.io/system-os_release.ID=centos
                    feature.node.kubernetes.io/system-os_release.VERSION_ID=7
                    feature.node.kubernetes.io/system-os_release.VERSION_ID.major=7
                    kubernetes.io/arch=amd64
                    kubernetes.io/hostname=master-10.20.31.105
                    kubernetes.io/os=linux
                    node-role.kubernetes.io/control-plane=
                    node-role.kubernetes.io/master=
                    node-role.kubernetes.io/worker=
                    node.kubernetes.io/exclude-from-external-load-balancers=
Annotations:        flannel.alpha.coreos.com/backend-data: {"VtepMAC":"c6:fb:4b:8a:bb:12"}
                    flannel.alpha.coreos.com/backend-type: vxlan
                    flannel.alpha.coreos.com/kube-subnet-manager: true
                    flannel.alpha.coreos.com/public-ip: 10.20.31.105
                    kubeadm.alpha.kubernetes.io/cri-socket: /var/run/dockershim.sock
                    nfd.node.kubernetes.io/feature-labels:
                      cpu-cpuid.ADX,cpu-cpuid.AESNI,cpu-cpuid.AVX,cpu-cpuid.AVX2,cpu-cpuid.AVX512BW,cpu-cpuid.AVX512CD,cpu-cpuid.AVX512DQ,cpu-cpuid.AVX512F,cpu-...
                    nfd.node.kubernetes.io/master.version: v0.14.2
                    nfd.node.kubernetes.io/worker.version: v0.14.2
                    node.alpha.kubernetes.io/ttl: 0
                    volumes.kubernetes.io/controller-managed-attach-detach: true
CreationTimestamp:  Tue, 12 Mar 2024 21:01:31 -0400

4、元件應用場景

Node Feature Discovery(NFD)元件的主要應用場景是在Kubernetes叢集中提供更智慧的節點排程。以下是一些NFD的常見應用場景:

  1. 智慧節點排程:NFD可以幫助Kubernetes排程器更好地瞭解節點的特性和資源,從而更智慧地選擇最適合執行特定工作負載的節點。例如,如果某個Pod需要較強的GPU支援,排程器可以利用NFD標籤來選擇具有適當GPU型號的節點。

  2. 資源約束和最佳化:透過將節點的特效能力以標籤的形式暴露給Kubernetes排程器,叢集管理員可以更好地理解和利用叢集中節點的資源情況,從而更好地進行資源約束和最佳化。

  3. 硬體感知的工作負載排程:對於特定的工作負載,可能需要特定型別或配置的硬體。NFD可以使排程器能夠更加智慧地選擇具有適當硬體特性的節點來執行這些工作負載。

  4. 叢集擴充套件性和效能:透過更智慧地分配工作負載到節點,NFD可以提高叢集的整體效能和效率。它可以幫助避免資源浪費,並確保工作負載能夠充分利用可用的硬體資源。

  5. 叢集自動化:NFD可以整合到自動化流程中,例如自動化部署或縮放工作負載。透過使用NFD,自動化系統可以更好地瞭解節點的特性和資源,從而更好地執行相應的操作。

總的來說,Node Feature Discovery(NFD)可以幫助提高Kubernetes叢集的智慧程度,使其能夠更好地適應各種型別的工作負載和節點特性,從而提高叢集的效能、可靠性和效率。

5、總結

如果您的 Kubernetes 叢集需要根據節點的硬體特性進行智慧排程或者對節點的硬體資源進行感知和利用,那麼安裝 Node Feature Discovery(NFD)是有必要的。然而,如果您的叢集中的節點都具有相似的硬體配置,且不需要考慮硬體資源的差異,那麼不需要安裝 NFD。

相關文章