Accelerating TLS Handshakes using Private Key Provider in Envoy

TLS operations can be accelerated or the private key can be protected using specialized hardware. This can be leveraged in Envoy using Envoy Private Key Provider is added to Envoy.

Today, there are two private key providers implemented in Envoy as contrib extensions:

Both of them are used to accelerate the TLS handshake through the hardware capabilities.

This task will walk you through the steps required to configure TLS Termination mode for TCP traffic while also using the Envoy Private Key Provider to accelerate the TLS handshake by leveraging QAT and the HW accelerator available on Intel SPR/EMR Xeon server platforms.

Prerequisites

For QAT

  • Install Linux kernel 5.17 or similar

  • Ensure the node has QAT devices by checking the QAT physical function devices presented. Supported Devices

    echo `(lspci -d 8086:4940 && lspci -d 8086:4941 && lspci -d 8086:4942 && lspci -d 8086:4943 && lspci -d 8086:4946 && lspci -d 8086:4947) | wc -l` supported devices found.
    
  • Enable IOMMU from BIOS

  • Enable IOMMU for Linux kernel

    Figure out the QAT VF device id

    lspci -d 8086:4941 && lspci -d 8086:4943 && lspci -d 8086:4947
    

    Attach the QAT device to vfio-pci through kernel parameter by the device id gotten from previous command.

    cat /etc/default/grub:
    GRUB_CMDLINE_LINUX="intel_iommu=on vfio-pci.ids=[QAT device id]"
    update-grub
    reboot
    

    Once the system is rebooted, check if the IOMMU has been enabled via the following command:

    dmesg| grep IOMMU
    [    1.528237] DMAR: IOMMU enabled
    
  • Enable virtual function devices for QAT device

    modprobe vfio_pci
    rmmod qat_4xxx
    modprobe qat_4xxx
    qat_device=$(lspci -D -d :[QAT device id] | awk '{print $1}')
    for i in $qat_device; do echo 16|sudo tee /sys/bus/pci/devices/$i/sriov_numvfs; done
    chmod a+rw /dev/vfio/*
    
  • Increase the container runtime memory lock limit (using the containerd as example here)

    mkdir /etc/systemd/system/containerd.service.d
    cat <<EOF >>/etc/systemd/system/containerd.service.d/memlock.conf
    [Service]
    LimitMEMLOCK=134217728
    EOF
    

    Restart the container runtime (for containerd, CRIO has similar concept)

    systemctl daemon-reload
    systemctl restart containerd
    
  • Install Intel® QAT Device Plugin for Kubernetes

    kubectl apply -k 'https://github.com/intel/intel-device-plugins-for-kubernetes/deployments/qat_plugin?ref=main'
    

    Verification of the plugin deployment and detection of QAT hardware can be confirmed by examining the resource allocations on the nodes:

    kubectl get node -o yaml| grep qat.intel.com
    

For CryptoMB:

It required the node with 3rd generation Intel Xeon Scalable processor server processors, or later.

  • For kubernetes Cluster, if not all nodes that support Intel® AVX-512 in Kubernetes cluster, you need to add some labels to divide these two kinds of nodes manually or using NFD.

    kubectl apply -k https://github.com/kubernetes-sigs/node-feature-discovery/deployment/overlays/default?ref=v0.15.1
    
  • Checking the available nodes with required cpu instructions:

    • Check the node labels if using NFD:

      kubectl get nodes -l feature.node.kubernetes.io/cpu-cpuid.AVX512F,feature.node.kubernetes.io/cpu-cpuid.AVX512DQ,feature.node.kubernetes.io/cpu-cpuid.AVX512BW,feature.node.kubernetes.io/cpu-cpuid.AVX512VBMI2,feature.node.kubernetes.io/cpu-cpuid.AVX512IFMA
      
    • Check CPUIDS manually on the node if without using NFD:

      cat /proc/cpuinfo |grep avx512f|grep avx512dq|grep avx512bw|grep avx512_vbmi2|grep avx512ifma
      

Installation

  • Follow the steps from the Quickstart to install Envoy Gateway.

  • Enable the EnvoyPatchPolicy feature, which will allow us to directly configure the Private Key Provider Envoy Filter, since Envoy Gateway does not directly expose this functionality.

cat <<EOF | kubectl apply -f -
apiVersion: v1
kind: ConfigMap
metadata:
  name: envoy-gateway-config
  namespace: envoy-gateway-system
data:
  envoy-gateway.yaml: |
    apiVersion: gateway.envoyproxy.io/v1alpha1
    kind: EnvoyGateway
    gateway:
      controllerName: gateway.envoyproxy.io/gatewayclass-controller
    extensionApis:
      enableEnvoyPatchPolicy: true
EOF

Save and apply the following resource to your cluster:

---
apiVersion: v1
kind: ConfigMap
metadata:
  name: envoy-gateway-config
  namespace: envoy-gateway-system
data:
  envoy-gateway.yaml: |
    apiVersion: gateway.envoyproxy.io/v1alpha1
    kind: EnvoyGateway
    gateway:
      controllerName: gateway.envoyproxy.io/gatewayclass-controller
    extensionApis:
      enableEnvoyPatchPolicy: true    
  • After updating the ConfigMap, you will need to restart the envoy-gateway deployment so the configuration kicks in

    kubectl rollout restart deployment envoy-gateway -n envoy-gateway-system
    

Create gateway for TLS termination

  • Follow the instructions in TLS Termination for TCP to setup a TCP gateway to terminate the TLS connection.

  • Update GatewayClass for using the envoyproxy image with contrib extensions and requests required resources.

cat <<EOF | kubectl apply -f -
apiVersion: gateway.networking.k8s.io/v1
kind: GatewayClass
metadata:
  name: eg
spec:
  controllerName: gateway.envoyproxy.io/gatewayclass-controller
  parametersRef:
    group: gateway.envoyproxy.io
    kind: EnvoyProxy
    name: custom-proxy-config
    namespace: envoy-gateway-system
EOF

Save and apply the following resource to your cluster:

---
apiVersion: gateway.networking.k8s.io/v1
kind: GatewayClass
metadata:
  name: eg
spec:
  controllerName: gateway.envoyproxy.io/gatewayclass-controller
  parametersRef:
    group: gateway.envoyproxy.io
    kind: EnvoyProxy
    name: custom-proxy-config
    namespace: envoy-gateway-system

Change EnvoyProxy configuration for QAT

Using the envoyproxy image with contrib extensions and add qat resources requesting, ensure the k8s scheduler find out a machine with required resource.

cat <<EOF | kubectl apply -f -
apiVersion: gateway.envoyproxy.io/v1alpha1
kind: EnvoyProxy
metadata:
  name: custom-proxy-config
  namespace: envoy-gateway-system
spec:
  concurrency: 1
  provider:
    type: Kubernetes
    kubernetes:
      envoyService:
        type: NodePort
      envoyDeployment:
        container:
          image: envoyproxy/envoy-contrib-dev:latest
          resources:
            requests:
              cpu: 1000m
              memory: 4096Mi
              qat.intel.com/cy: '1'
            limits:
              cpu: 1000m
              memory: 4096Mi
              qat.intel.com/cy: '1'
EOF

Save and apply the following resource to your cluster:

---
apiVersion: gateway.envoyproxy.io/v1alpha1
kind: EnvoyProxy
metadata:
  name: custom-proxy-config
  namespace: envoy-gateway-system
spec:
  concurrency: 1
  provider:
    type: Kubernetes
    kubernetes:
      envoyService:
        type: NodePort
      envoyDeployment:
        container:
          image: envoyproxy/envoy-contrib-dev:latest
          resources:
            requests:
              cpu: 1000m
              memory: 4096Mi
              qat.intel.com/cy: '1'
            limits:
              cpu: 1000m
              memory: 4096Mi
              qat.intel.com/cy: '1'

Change EnvoyProxy configuration for CryptoMB

Using the envoyproxy image with contrib extensions and add the node affinity to scheduling the Envoy Gateway pod on the machine with required CPU instructions.

cat <<EOF | kubectl apply -f -
apiVersion: gateway.envoyproxy.io/v1alpha1
kind: EnvoyProxy
metadata:
  name: custom-proxy-config
  namespace: envoy-gateway-system
spec:
  concurrency: 1
  provider:
    type: Kubernetes
    kubernetes:
      envoyService:
        type: NodePort
      envoyDeployment:
        container:
          image: envoyproxy/envoy-contrib-dev:latest
          resources:
            requests:
              cpu: 1000m
              memory: 4096Mi
            limits:
              cpu: 1000m
              memory: 4096Mi
        pod:
          affinity:
            nodeAffinity:
              requiredDuringSchedulingIgnoredDuringExecution:
                nodeSelectorTerms:
                - matchExpressions:
                  - key: feature.node.kubernetes.io/cpu-cpuid.AVX512F
                    operator: Exists
                  - key: feature.node.kubernetes.io/cpu-cpuid.AVX512DQ
                    operator: Exists
                  - key: feature.node.kubernetes.io/cpu-cpuid.AVX512BW
                    operator: Exists
                  - key: feature.node.kubernetes.io/cpu-cpuid.AVX512IFMA
                    operator: Exists
                  - key: feature.node.kubernetes.io/cpu-cpuid.AVX512VBMI2
                    operator: Exists
EOF

Save and apply the following resource to your cluster:

---
apiVersion: gateway.envoyproxy.io/v1alpha1
kind: EnvoyProxy
metadata:
  name: custom-proxy-config
  namespace: envoy-gateway-system
spec:
  concurrency: 1
  provider:
    type: Kubernetes
    kubernetes:
      envoyService:
        type: NodePort
      envoyDeployment:
        container:
          image: envoyproxy/envoy-contrib-dev:latest
          resources:
            requests:
              cpu: 1000m
              memory: 4096Mi
            limits:
              cpu: 1000m
              memory: 4096Mi
        pod:
          affinity:
            nodeAffinity:
              requiredDuringSchedulingIgnoredDuringExecution:
                nodeSelectorTerms:
                - matchExpressions:
                  - key: feature.node.kubernetes.io/cpu-cpuid.AVX512F
                    operator: Exists
                  - key: feature.node.kubernetes.io/cpu-cpuid.AVX512DQ
                    operator: Exists
                  - key: feature.node.kubernetes.io/cpu-cpuid.AVX512BW
                    operator: Exists
                  - key: feature.node.kubernetes.io/cpu-cpuid.AVX512IFMA
                    operator: Exists
                  - key: feature.node.kubernetes.io/cpu-cpuid.AVX512VBMI2
                    operator: Exists

Or using preferredDuringSchedulingIgnoredDuringExecution for best effort scheduling, or not doing any node affinity, just doing the random scheduling. The CryptoMB private key provider supports software fallback if the required CPU instructions aren’t here.

Apply EnvoyPatchPolicy to enable private key provider

Benchmark before enabling private key provider

First follow the instructions in TLS Termination for TCP to do the functionality test.

Ensure the cpu frequency governor set as performance.

export NUM_CPUS=`lscpu | grep "^CPU(s):"|awk '{print $2}'`
for i in `seq 0 1 $NUM_CPUS`; do sudo cpufreq-set -c $i -g performance; done

Using the nodeport as the example, fetch the node port from envoy gateway service.

export ENVOY_SERVICE=$(kubectl get svc -n envoy-gateway-system --selector=gateway.envoyproxy.io/owning-gateway-namespace=default,gateway.envoyproxy.io/owning-gateway-name=eg -o jsonpath='{.items[0].metadata.name}')
export NODE_PORT=$(kubectl -n envoy-gateway-system get svc/$ENVOY_SERVICE -o jsonpath='{.spec.ports[0].nodePort}')
echo "127.0.0.1 www.example.com" >> /etc/hosts

Benchmark the gateway with fortio.

fortio load -c 10 -k -qps 0 -t 30s -keepalive=false https://www.example.com:${NODE_PORT}

For QAT

cat <<EOF | kubectl apply -f -
apiVersion: gateway.envoyproxy.io/v1alpha1
kind: EnvoyPatchPolicy
metadata:
  name: key-provider-patch-policy
  namespace: default
spec:
  targetRef:
    group: gateway.networking.k8s.io
    kind: Gateway
    name: eg
  type: JSONPatch
  jsonPatches:
    - type: "type.googleapis.com/envoy.extensions.transport_sockets.tls.v3.Secret"
      name: default/example-cert
      operation:
        op: add
        path: "/tls_certificate/private_key_provider"
        value:
          provider_name: qat
          typed_config:
            "@type": "type.googleapis.com/envoy.extensions.private_key_providers.qat.v3alpha.QatPrivateKeyMethodConfig"
            private_key:
              inline_string: |
                abcd
            poll_delay: 0.001s
          fallback: true
    - type: "type.googleapis.com/envoy.extensions.transport_sockets.tls.v3.Secret"
      name: default/example-cert
      operation:
        op: copy
        from: "/tls_certificate/private_key"
        path: "/tls_certificate/private_key_provider/typed_config/private_key"
EOF

Save and apply the following resource to your cluster:

---
apiVersion: gateway.envoyproxy.io/v1alpha1
kind: EnvoyPatchPolicy
metadata:
  name: key-provider-patch-policy
  namespace: default
spec:
  targetRef:
    group: gateway.networking.k8s.io
    kind: Gateway
    name: eg
  type: JSONPatch
  jsonPatches:
    - type: "type.googleapis.com/envoy.extensions.transport_sockets.tls.v3.Secret"
      name: default/example-cert
      operation:
        op: add
        path: "/tls_certificate/private_key_provider"
        value:
          provider_name: qat
          typed_config:
            "@type": "type.googleapis.com/envoy.extensions.private_key_providers.qat.v3alpha.QatPrivateKeyMethodConfig"
            private_key:
              inline_string: |
                abcd                
            poll_delay: 0.001s
          fallback: true
    - type: "type.googleapis.com/envoy.extensions.transport_sockets.tls.v3.Secret"
      name: default/example-cert
      operation:
        op: copy
        from: "/tls_certificate/private_key"
        path: "/tls_certificate/private_key_provider/typed_config/private_key"

For CryptoMB

cat <<EOF | kubectl apply -f -
apiVersion: gateway.envoyproxy.io/v1alpha1
kind: EnvoyPatchPolicy
metadata:
  name: key-provider-patch-policy
  namespace: default
spec:
  targetRef:
    group: gateway.networking.k8s.io
    kind: Gateway
    name: eg
  type: JSONPatch
  jsonPatches:
    - type: "type.googleapis.com/envoy.extensions.transport_sockets.tls.v3.Secret"
      name: default/example-cert
      operation:
        op: add
        path: "/tls_certificate/private_key_provider"
        value:
          provider_name: cryptomb
          typed_config:  
            "@type": "type.googleapis.com/envoy.extensions.private_key_providers.cryptomb.v3alpha.CryptoMbPrivateKeyMethodConfig"
            private_key:
              inline_string: |
                abcd
            poll_delay: 0.001s
          fallback: true
    - type: "type.googleapis.com/envoy.extensions.transport_sockets.tls.v3.Secret"
      name: default/example-cert
      operation:
        op: copy
        from: "/tls_certificate/private_key"
        path: "/tls_certificate/private_key_provider/typed_config/private_key"
EOF

Save and apply the following resource to your cluster:

---
apiVersion: gateway.envoyproxy.io/v1alpha1
kind: EnvoyPatchPolicy
metadata:
  name: key-provider-patch-policy
  namespace: default
spec:
  targetRef:
    group: gateway.networking.k8s.io
    kind: Gateway
    name: eg
  type: JSONPatch
  jsonPatches:
    - type: "type.googleapis.com/envoy.extensions.transport_sockets.tls.v3.Secret"
      name: default/example-cert
      operation:
        op: add
        path: "/tls_certificate/private_key_provider"
        value:
          provider_name: cryptomb
          typed_config:  
            "@type": "type.googleapis.com/envoy.extensions.private_key_providers.cryptomb.v3alpha.CryptoMbPrivateKeyMethodConfig"
            private_key:
              inline_string: |
                abcd                
            poll_delay: 0.001s
          fallback: true
    - type: "type.googleapis.com/envoy.extensions.transport_sockets.tls.v3.Secret"
      name: default/example-cert
      operation:
        op: copy
        from: "/tls_certificate/private_key"
        path: "/tls_certificate/private_key_provider/typed_config/private_key"

Benchmark after enabling private key provider

First follow the instructions in TLS Termination for TCP to do the functionality test again.

Benchmark the gateway with fortio.

fortio load -c 64 -k -qps 0 -t 30s -keepalive=false https://www.example.com:${NODE_PORT}

You will see a performance boost after private key provider enabled. For example, you will get results as below.

Without private key provider:

All done 43069 calls (plus 10 warmup) 6.966 ms avg, 1435.4 qps

With CryptoMB private key provider, the QPS is over 2 times than without private key provider.

All done 93983 calls (plus 128 warmup) 40.880 ms avg, 3130.5 qps

With QAT private key provider, the QPS is over 3 times than without private key provider

All done 134746 calls (plus 128 warmup) 28.505 ms avg, 4489.6 qps

Last modified December 20, 2024: chore: fix typo (#4958) (2a10d47)