//   have observed 12 micros on 64-bit linux systems to wake up a parked thread). So if the
  //   timeout is small we shouldn't park(). This needs to be traded off with the cpu overhead of
  //   spinning, so we use SPIN_THRESHOLD_NANOS which is what AbstractQueuedSynchronizer uses for
  //   similar purposes.
  // * We want to behave reasonably for timeouts of 0

- A new kubelet command line option, `--reserved-cpus`, is introduced to explicitly define the CPU list that will be reserved for system. For example, if `--reserved-cpus=0,1,2,3` is specified, then cpu 0,1,2,3 will be reserved for the system.  On a system with 24 CPUs, the user may specify `isolcpus=4-23` for the kernel option and use CPU 4-23 for the user containers. ([#83592](https://github.com/kubernetes/kubernetes/pull/83592), [@jianzzha](https://github.com/jianzzha))...

  //   have observed 12 micros on 64-bit linux systems to wake up a parked thread). So if the
  //   timeout is small we shouldn't park(). This needs to be traded off with the cpu overhead of
  //   spinning, so we use SPIN_THRESHOLD_NANOS which is what AbstractQueuedSynchronizer uses for
  //   similar purposes.
  // * We want to behave reasonably for timeouts of 0

- The CPU Manager will now validate the state of the node,  enabling Kubernetes to maintain the CPU topology even if resources change. ([#66718](https://github.com/kubernetes/kubernetes/pull/66718), [@ipuustin](https://github.com/ipuustin))

 * either of these extremes, {@code Striped} allows the user to trade between required concurrency
 * and memory footprint. For example, if a set of tasks are CPU-bound, one could easily create a
 * very compact {@code Striped<Lock>} of {@code availableProcessors() * 4} stripes, instead of
 * possibly thousands of locks which could be created in a {@code Map<K, Lock>} structure.
 *

The Resource Management Working Group graduated three features to beta in the 1.10 release.  First, [CPU Manager](https://kubernetes.io/docs/tasks/administer-cluster/cpu-management-policies/), which allows users to request exclusive CPU cores.  This helps performance in a variety of use-cases, including network latency sensitive applications, as well as applications that benefit from CPU cache residency.  Next, [Huge Pages](https://kubernetes.io/docs/tasks/manage-hugepages/scheduling-hugepages/),...

Безусловно, на этом же сервере будут работать и **другие процессы**, которые не относятся к вашему приложению.

Интересная деталь заключается в том, что процент **использования центрального процессора (CPU)** каждым процессом может сильно меняться с течением времени, но объём занимаемой **оперативной памяти (RAM)** остаётся относительно **стабильным**.

          "interval": "",
          "legendFormat": "{{server}}",
          "range": true,
          "refId": "A"
        }
      ],
      "title": "CPU Usage",
      "type": "gauge"
    },
    {
      "datasource": {
        "type": "prometheus",
        "uid": "${DS_PROMETHEUS}"
      },

- Fixed relative CPU priority for pods where containers explicitly request zero cpu by giving the lowest priority instead of falling back to the cpu limit to avoid possible cpu starvation of other pods. ([#108832](https://github.com/kubernetes/kubernetes/pull/108832), [@waynepeking348](https://github.com/waynepeking348))

	resp, err := serverInfoRPC.Call(ctx, client.gridConn(), grid.NewMSSWith(map[string]string{peerRESTMetrics: strconv.FormatBool(metrics)}))
	return resp.ValueOrZero(), err
}

// GetCPUs - fetch CPU information for a remote node.
func (client *peerRESTClient) GetCPUs(ctx context.Context) (info madmin.CPUs, err error) {
	resp, err := getCPUsHandler.Call(ctx, client.gridConn(), grid.NewMSS())
	return resp.ValueOrZero(), err