### Memória por Processo { #memory-per-process }

Agora, quando o programa carrega coisas na memória, por exemplo, um modelo de aprendizado de máquina em uma variável, ou o conteúdo de um arquivo grande em uma variável, tudo isso **consome um pouco da memória (RAM)** do servidor.

 *       block indefinitely or fail if the source creates an infinite reader.
 *   <li><b>Non-destructive:</b> A <i>destructive</i> reader will consume or otherwise alter the
 *       source as they are read from it. A source that provides such readers will not be reusable,
 *       and operations that read from the stream (including {@link #length()}, in some

 *       block indefinitely or fail if the source creates an infinite reader.
 *   <li><b>Non-destructive:</b> A <i>destructive</i> reader will consume or otherwise alter the
 *       source as they are read from it. A source that provides such readers will not be reusable,
 *       and operations that read from the stream (including {@link #length()}, in some

            // When
            int bytesConsumed = fileFsFullSizeInfo.decode(buffer, 0, buffer.length);

            // Then
            assertEquals(32, bytesConsumed); // Should always consume exactly 32 bytes
        }

        @Test
        @DisplayName("Should decode from middle of buffer correctly")
        void shouldDecodeFromMiddleOfBuffer() throws SMBProtocolDecodingException {

## Más motivación { #more-motivation }

Revisa esta función, ya tiene anotaciones de tipos:

{* ../../docs_src/python_types/tutorial003_py39.py hl[1] *}

Porque el editor conoce los tipos de las variables, no solo obtienes autocompletado, también obtienes chequeo de errores:

<img src="/img/python-types/image04.png">

Ahora sabes que debes corregirlo, convertir `age` a un string con `str(age)`:

### Memoria por Proceso { #memory-per-process }

Ahora, cuando el programa carga cosas en memoria, por ejemplo, un modelo de Machine Learning en una variable, o el contenido de un archivo grande en una variable, todo eso **consume un poco de la memoria (RAM)** del servidor.

 *       block indefinitely or fail if the source creates an infinite stream.
 *   <li><b>Non-destructive:</b> A <i>destructive</i> stream will consume or otherwise alter the
 *       bytes of the source as they are read from it. A source that provides such streams will not
 *       be reusable, and operations that read from the stream (including {@link #size()}, in some

   * @return a view of the supplied iterable that wraps each generated iterator through {@link
   *     Iterators#consumingIterator(Iterator)}; for queues, an iterable that generates iterators
   *     that return and consume the queue's elements in queue order
   * @see Iterators#consumingIterator(Iterator)
   * @since 2.0
   */
  public static <T extends @Nullable Object> Iterable<T> consumingIterable(Iterable<T> iterable) {

    * New docs: [FastAPI CLI](https://fastapi.tiangolo.com/fastapi-cli/).

Try it out with:

```console
$ pip install --upgrade fastapi

$ fastapi dev main.py


 ╭────────── FastAPI CLI - Development mode ───────────╮
 │                                                     │

* une opération effectuée à distance par une API se termine
* une opération en BDD se termine
* une requête à une BDD renvoie un résultat
* etc.

Le temps d'exécution étant consommé majoritairement par l'attente d'opérations <abbr title="Input/Output ou Entrées et Sorties ">I/O</abbr> on appelle ceci des opérations <a href="https://fr.wikipedia.org/wiki/I/O_bound" class="external-link" target="_blank">"I/O bound"</a>.