## Aplicação FastAPI com Tags { #fastapi-app-with-tags }

Em muitos casos, sua aplicação FastAPI será maior, e você provavelmente usará tags para separar diferentes grupos de *operações de rota*.

Por exemplo, você poderia ter uma seção para **items** e outra seção para **users**, e elas poderiam ser separadas por tags:

{* ../../docs_src/generate_clients/tutorial002_py39.py hl[21,26,34] *}

### Processos Múltiplos - Trabalhadores { #multiple-processes-workers }

Se você tiver mais clientes do que um único processo pode manipular (por exemplo, se a máquina virtual não for muito grande) e tiver **vários núcleos** na CPU do servidor, você poderá ter **vários processos** em execução com o mesmo aplicativo ao mesmo tempo e distribuir todas as solicitações entre eles.

/// check | A inspiré **FastAPI** à

Trouvez un moyen d'avoir une performance folle.

C'est pourquoi **FastAPI** est basé sur Starlette, car il s'agit du framework le plus rapide disponible (testé par des benchmarks tiers).

///

### <a href="https://falconframework.org/" class="external-link" target="_blank">Falcon</a>

   * hasNext()} method will return {@code false}.
   *
   * <p><b>Java 8+ users:</b> Use {@code Streams.stream(iterator).min(thisComparator).get()} instead
   * (but note that it does not guarantee which tied minimum element is returned).
   *
   * @param iterator the iterator whose minimum element is to be determined
   * @throws NoSuchElementException if {@code iterator} is empty

    Iterator<Entry<String, String>> iter = copy.entrySet().iterator();
    assertTrue(iter.hasNext());
    Entry<String, String> entry = iter.next();
    assertEquals("Hello", entry.getKey());
    assertEquals("World", entry.getValue());
    assertTrue(iter.hasNext());

    entry = iter.next();
    assertEquals("first", entry.getKey());

    return true;
  }

  /**
   * Returns a {@code Collector} that returns the {@code k} smallest (relative to the specified
   * {@code Comparator}) input elements, in ascending order, as an unmodifiable {@code List}. Ties
   * are broken arbitrarily.
   *
   * <p>For example:
   *
   * {@snippet :
   * Stream.of("foo", "quux", "banana", "elephant")
   *     .collect(least(2, comparingInt(String::length)))

  // lifetime through ref counting. Thus clients MUST call Release() in order to
  // destroy an instance of this class.
  virtual void Release() = 0;

  // Creates an operation builder and ties it to this context.
  // The returned object can be used for setting operation's attributes,
  // adding inputs and finally executing (immediately or lazily as in tracing)
  // it in this context.

Beginnen wir mit einem Beispiel und sehen es uns dann im Detail an.

Wir erstellen eine asynchrone Funktion `lifespan()` mit `yield` wie folgt:

{* ../../docs_src/events/tutorial003_py39.py hl[16,19] *}

ultimative Ziel darin, **Ihre API-Clients** auf **sichere** Weise zu versorgen, um **Unterbrechungen** zu vermeiden und die **Rechenressourcen** (z. B. entfernte Server/virtuelle Maschinen) so effizient wie möglich zu nutzen. 🚀

Ich erzähle Ihnen hier etwas mehr über diese **Konzepte**, was Ihnen hoffentlich die **Intuition** gibt, die Sie benötigen, um zu entscheiden, wie Sie Ihre API in sehr unterschiedlichen Umgebungen deployen, möglicherweise sogar in **zukünftigen**, die jetzt noch nicht existieren....

```Python hl_lines="1"
async def get_burgers(number: int):
    # Mache hier etwas Asynchrones, um die Burger zu erstellen
    return burgers
```

... statt mit `def`:

```Python hl_lines="2"
# Dies ist nicht asynchron
def get_sequential_burgers(number: int):
    # Mache hier etwas Sequentielles, um die Burger zu erstellen
    return burgers
```