---
title: 资源指标管道
content_type: concept
---
<!--
reviewers:
- fgrzadkowski
- piosz
title: Resource metrics pipeline
content_type: concept
-->

<!-- overview -->

<!--
Resource usage metrics, such as container CPU and memory usage,
are available in Kubernetes through the Metrics API. These metrics can be accessed either directly
by the user with the `kubectl top` command, or by a controller in the cluster, for example
Horizontal Pod Autoscaler, to make decisions.
-->
资源使用指标，例如容器 CPU 和内存使用率，可通过 Metrics API 在 Kubernetes 中获得。
这些指标可以直接被用户访问，比如使用 `kubectl top` 命令行，或者被集群中的控制器
（例如 Horizontal Pod Autoscalers) 使用来做决策。

<!-- body -->

<!--
## The Metrics API

Through the Metrics API, you can get the amount of resource currently used
by a given node or a given pod. This API doesn't store the metric values,
so it's not possible, for example, to get the amount of resources used by a
given node 10 minutes ago.
-->
## Metrics API   {#the-metrics-api}

通过 Metrics API，你可以获得指定节点或 Pod 当前使用的资源量。
此 API 不存储指标值，因此想要获取某个指定节点 10 分钟前的
资源使用量是不可能的。

<!--
The API is no different from any other API:
-->
此 API 与其他 API 没有区别：

<!--
- it is discoverable through the same endpoint as the other Kubernetes APIs under the path: `/apis/metrics.k8s.io/`
- it offers the same security, scalability, and reliability guarantees
-->
- 此 API 和其它 Kubernetes API 一起位于同一端点（endpoint）之下且可发现，
  路径为 `/apis/metrics.k8s.io/` 
- 它具有相同的安全性、可扩展性和可靠性保证

<!--
The API is defined in [k8s.io/metrics](https://github.com/kubernetes/metrics/blob/master/pkg/apis/metrics/v1beta1/types.go)
repository. You can find more information about the API there.
-->
Metrics API 在 [k8s.io/metrics](https://github.com/kubernetes/metrics/blob/master/pkg/apis/metrics/v1beta1/types.go)
仓库中定义。你可以在那里找到有关 Metrics API 的更多信息。

<!--
The API requires metrics server to be deployed in the cluster. Otherwise it will be not available.
-->
{{< note >}}
Metrics API 需要在集群中部署 Metrics Server。否则它将不可用。
{{< /note >}}

<!--
## Measuring Resource Usage

### CPU

CPU is reported as the average usage, in
[CPU cores](/docs/concepts/configuration/manage-resources-containers/#meaning-of-cpu),
over a period of time. This value is derived by taking a rate over a cumulative CPU counter
provided by the kernel (in both Linux and Windows kernels).
The kubelet chooses the window for the rate calculation.
-->
## 度量资源用量   {#measuring-resource-usage}

### CPU

CPU 用量按其一段时间内的平均值统计，单位为
[CPU 核](/zh/docs/concepts/configuration/manage-resources-containers/#meaning-of-cpu)。
此度量值通过在内核（包括 Linux 和 Windows）提供的累积 CPU 计数器乘以一个系数得到。
`kubelet` 组件负责选择计算系数所使用的窗口大小。

<!--
### Memory

Memory is reported as the working set, in bytes, at the instant the metric was collected.
In an ideal world, the "working set" is the amount of memory in-use that cannot be freed under memory pressure.
However, calculation of the working set varies by host OS, and generally makes heavy use of heuristics to produce an estimate.
It includes all anonymous (non-file-backed) memory since Kubernetes does not support swap.
The metric typically also includes some cached (file-backed) memory, because the host OS cannot always reclaim such pages.
-->
### 内存  {#memory}

内存用量按工作集（Working Set）的大小字节数统计，其数值为收集度量值的那一刻的内存用量。
如果一切都很理想化，“工作集” 是任务在使用的内存总量，该内存是不可以在内存压力较大
的情况下被释放的。
不过，具体的工作集计算方式取决于宿主 OS，有很大不同，且通常都大量使用启发式
规则来给出一个估计值。
其中包含所有匿名内存使用（没有后台文件提供存储者），因为 Kubernetes 不支持交换分区。
度量值通常包含一些高速缓存（有后台文件提供存储）内存，因为宿主操作系统并不是总能
回收这些页面。

<!--
## Metrics Server

[Metrics Server](https://github.com/kubernetes-sigs/metrics-server) is a cluster-wide aggregator of resource usage data.
By default, it is deployed in clusters created by `kube-up.sh` script
as a Deployment object. If you use a different Kubernetes setup mechanism you can deploy it using the provided
[deployment components.yaml](https://github.com/kubernetes-sigs/metrics-server/releases) file.
-->
## Metrics 服务器    {#metrics-server}

[Metrics 服务器](https://github.com/kubernetes-sigs/metrics-server)
是集群范围资源用量数据的聚合器。
默认情况下，在由 `kube-up.sh` 脚本创建的集群中会以 Deployment 的形式被部署。
如果你使用其他 Kubernetes 安装方法，则可以使用提供的
[部署组件 components.yaml](https://github.com/kubernetes-sigs/metrics-server/releases)
来部署。

<!--
Metric server collects metrics from the Summary API, exposed by
[Kubelet](/docs/reference/command-line-tools-reference/kubelet/) on each node, and is registered with the main API server via
[Kubernetes aggregator](/docs/concepts/extend-kubernetes/api-extension/apiserver-aggregation/).
-->
Metric 服务器从每个节点上的 [kubelet](/zh/docs/reference/command-line-tools-reference/kubelet/)
公开的 Summary API 中采集指标信息。
该 API 通过
[Kubernetes 聚合器](/zh/docs/concepts/extend-kubernetes/api-extension/apiserver-aggregation/)
注册到主 API 服务器上。

<!--
Learn more about the metrics server in
[the design doc](https://github.com/kubernetes/community/blob/master/contributors/design-proposals/instrumentation/metrics-server.md).
-->
在[设计文档](https://github.com/kubernetes/community/blob/master/contributors/design-proposals/instrumentation/metrics-server.md)
中可以了解到有关 Metrics 服务器的更多信息。

<!--
### Summary API Source

The [Kubelet](/docs/reference/command-line-tools-reference/kubelet/) gathers stats at node, volume, pod and container level, and emits
them in the [Summary API](https://github.com/kubernetes/kubernetes/blob/7d309e0104fedb57280b261e5677d919cb2a0e2d/staging/src/k8s.io/kubelet/pkg/apis/stats/v1alpha1/types.go)
for consumers to read.
-->

### 摘要 API 来源

[Kubelet](/zh/docs/reference/command-line-tools-reference/kubelet/) 在节点、卷、pod 和容器级别收集统计信息，
并在[摘要API](https://github.com/kubernetes/kubernetes/blob/7d309e0104fedb57280b261e5677d919cb2a0e2d/staging/src/k8s.io/kubelet/pkg/apis/stats/v1alpha1/types.go) 
中提供它们的统计信息供消费者阅读。

<!--
Pre-1.23, these resources have been primarily gathered from [cAdvisor](https://github.com/google/cadvisor). However, in 1.23 with the
introduction of the `PodAndContainerStatsFromCRI` FeatureGate, container and pod level stats can be gathered by the CRI implementation.
Note: this also requires support from the CRI implementations (containerd >= 1.6.0, CRI-O >= 1.23.0).
-->

在 1.23 版本前，这些资源主要来自 [cAdvisor](https://github.com/google/cadvisor)。但在 1.23 版本中
引入了 `PodAndContainerStatsFromCRI` FeatureGate，
CRI 实现了可以收集容器和 pod 级别的统计信息。
注意：这需要 CRI 实现的支持（containerd >= 1.6.0，CRI-O >= 1.23.0）。