issues with Docker 1.9.1 can be found below.
  * CPU hardcapping will be enabled by default for containers with CPU limit set,
if supported by the kernel. You should either adjust your CPU limit, or set CPU
request only, if you want to avoid hardcapping. If the kernel does not support
CPU Quota, NodeStatus will contain a warning indicating that CPU Limits cannot
be enforced.

		m := collectLocalMetrics(types, collectMetricsOpts{
			hosts: map[string]struct{}{
				globalLocalNodeName: {},
			},
		})

		for _, hm := range m.ByHost {
			if hm.CPU != nil {
				v = *hm.CPU
				break
			}
		}

		return v, err
	}

	return cachevalue.NewFromFunc(1*time.Minute,
		cachevalue.Opts{ReturnLastGood: true},
		loadCPUMetrics)
}

И так как основное время выполнения уходит на реальную работу (а не ожидание), а работу в компьютере выполняет <abbr title="Central Processing Unit – Центральный процессор">CPU</abbr>, такие задачи называют «ограниченными процессором» (CPU bound).

---

Типичные примеры CPU-bound операций — те, которые требуют сложной математической обработки.

Например:

* Обработка **аудио** или **изображений**.

TFCI_BUILD_PIP_PACKAGE_WHEEL_NAME_ARG="--repo_env=WHEEL_NAME=tensorflow"
TFCI_BUILD_PIP_PACKAGE_ADDITIONAL_WHEEL_NAMES="tensorflow_cpu"
TFCI_OUTPUT_DIR=build_output
TFCI_FIND_BIN=C:/tools/msys64/usr/bin/find.exe
TFCI_LIB_SUFFIX="-cpu-windows-x86_64"
# auditwheel is not supported for Windows
TFCI_WHL_AUDIT_ENABLE=0
TFCI_WHL_AUDIT_PLAT=0
# Tests are extremely slow at the moment
TFCI_WHL_BAZEL_TEST_ENABLE=0
TFCI_WHL_SIZE_LIMIT=450M

---

Typische Beispiele für CPU-lastige Vorgänge sind Dinge, die komplexe mathematische Berechnungen erfordern.

Zum Beispiel:

* **Audio-** oder **Bildbearbeitung**.

 * java.util.LinkedHashSet}. Generally speaking, this class reduces object allocation and memory
 * consumption at the price of moderately increased constant factors of CPU. Only use this class
 * when there is a specific reason to prioritize memory over CPU.
 *
 * @author Louis Wasserman
 */
@GwtIncompatible // not worth using in GWT for now
final class CompactLinkedHashSet<E extends @Nullable Object> extends CompactHashSet<E> {

 * java.util.LinkedHashSet}. Generally speaking, this class reduces object allocation and memory
 * consumption at the price of moderately increased constant factors of CPU. Only use this class
 * when there is a specific reason to prioritize memory over CPU.
 *
 * @author Louis Wasserman
 */
@GwtIncompatible // not worth using in GWT for now
final class CompactLinkedHashSet<E extends @Nullable Object> extends CompactHashSet<E> {

          CI_DOCKER_BUILD_EXTRA_PARAMS="--build-arg py_major_minor_version=${{ matrix.pyver }} --build-arg is_nightly=${is_nightly} --build-arg tf_project_name=${tf_project_name}" \

 * java.util.LinkedHashMap}. Generally speaking, this class reduces object allocation and memory
 * consumption at the price of moderately increased constant factors of CPU. Only use this class
 * when there is a specific reason to prioritize memory over CPU.
 *
 * @author Louis Wasserman
 */
@J2ktIncompatible // no support for access-order mode in LinkedHashMap delegate
@GwtIncompatible // not worth using in GWT for now

 * java.util.LinkedHashMap}. Generally speaking, this class reduces object allocation and memory
 * consumption at the price of moderately increased constant factors of CPU. Only use this class
 * when there is a specific reason to prioritize memory over CPU.
 *
 * @author Louis Wasserman
 */
@J2ktIncompatible // no support for access-order mode in LinkedHashMap delegate
@GwtIncompatible // not worth using in GWT for now