* As connecting to an email server and sending an email tends to be "slow" (several seconds), you can return the response right away and send the email notification in the background.
* Processing data:
    * For example, let's say you receive a file that must go through a slow process, you can return a response of "Accepted" (HTTP 202) and process it in the background.

## Using `BackgroundTasks`

!!! tip
    As FastAPI is an async framework, one request could start being processed, and before finishing, another request could be received and start processing as well, and it all could be processed in the same thread.

That way, you don't have to read it all first in memory, and you can pass that generator function to the `StreamingResponse`, and return it.

This includes many libraries to interact with cloud storage, video processing, and others.

```{ .python .annotate hl_lines="2  10-12  14" }
{!../../../docs_src/custom_response/tutorial008.py!}
```

    And then we close it in the `finally` block.

    This way we make sure the database session is always closed after the request. Even if there was an exception while processing the request.

이 시나리오에서, (당신을 포함한) 각각의 청소부들은 프로세서가 될 것이고, 각자의 역할을 수행합니다.

실행 시간의 대부분이 대기가 아닌 실제 작업에 소요되고, 컴퓨터에서 작업은 <abbr title="Central Processing Unit">CPU</abbr>에서 이루어지므로, 이러한 문제를 "CPU에 묶였"다고 합니다.

---

CPU에 묶인 연산에 관한 흔한 예시는 복잡한 수학 처리를 필요로 하는 경우입니다.

예를 들어:

* **오디오** 또는 **이미지** 처리.

Und da die meiste Ausführungszeit durch tatsächliche Arbeit (anstatt durch Warten) in Anspruch genommen wird und die Arbeit in einem Computer von einer <abbr title="Central Processing Unit – Zentrale Recheneinheit">CPU</abbr> erledigt wird, werden diese Probleme als „CPU-lastig“ („CPU bound“) bezeichnet.

---

Y como la mayor parte del tiempo de ejecución lo coge el trabajo real (en lugar de esperar), y el trabajo en un sistema lo realiza una <abbr title = "Central Processing Unit. En español: Unidad Central de Procesamiento."> CPU </abbr>, a estos problemas se les llama "<abbr title="En español: atado a CPU.">CPU bound</abbr>".

---

👉 😐, 🔠 1️⃣ 🧹 (🔌 👆) 🔜 🕹, 🤸 👫 🍕 👨‍🏭.

&amp; 🏆 🛠️ 🕰 ✊ ☑ 👷 (↩️ ⌛), &amp; 👷 💻 ⌛ <abbr title="Central Processing Unit">💽</abbr>, 👫 🤙 👫 ⚠ "💽 🎁".

---

⚠ 🖼 💽 🔗 🛠️ 👜 👈 🚚 🏗 🧪 🏭.

🖼:

* **🎧** ⚖️ **🖼 🏭**.
* **💻 👓**: 🖼 ✍ 💯 🔅, 🔠 🔅 ✔️ 3️⃣ 💲 / 🎨, 🏭 👈 🛎 🚚 💻 🕳 🔛 📚 🔅, 🌐 🎏 🕰.

无论是否轮流执行（并发），都需要相同的时间来完成，而你也会完成相同的工作量。

但在这种情况下，如果你能带上 8 名前收银员/厨师，现在是清洁工一起清扫，他们中的每一个人（加上你）都能占据房子的一个区域来清扫，你就可以在额外的帮助下并行的更快地完成所有工作。

在这个场景中，每个清洁工（包括您）都将是一个处理器，完成这个工作的一部分。

由于大多数执行时间是由实际工作（而不是等待）占用的，并且计算机中的工作是由 <abbr title="Central Processing Unit">CPU</abbr> 完成的，所以他们称这些问题为"CPU 密集型"。

---

CPU 密集型操作的常见示例是需要复杂的数学处理。

例如：

* **音频**或**图像**处理；
* **计算机视觉**: 一幅图像由数百万像素组成，每个像素有3种颜色值，处理通常需要同时对这些像素进行计算；