## 서버 머신과 서버 프로그램 { #server-machine-and-server-program }

이름에 관해 기억해 둘 작은 디테일이 있습니다. 💡

"**server**"라는 단어는 보통 원격/클라우드 컴퓨터(물리 또는 가상 머신)와, 그 머신에서 실행 중인 프로그램(예: Uvicorn) 둘 다를 가리키는 데 사용됩니다.

일반적으로 "server"를 읽을 때, 이 두 가지 중 하나를 의미할 수 있다는 점을 기억하세요.

원격 머신을 가리킬 때는 **server**라고 부르는 것이 일반적이지만, **machine**, **VM**(virtual machine), **node**라고 부르기도 합니다. 이것들은 보통 Linux를 실행하는 원격 머신의 한 형태를 뜻하며, 그곳에서 프로그램을 실행합니다.

		}
	}

	return host, port, nil
}

// isLocalHost - checks if the given parameter
// correspond to one of the local IP of the
// current machine
func isLocalHost(host string, port string, localPort string) (bool, error) {
	hostIPs, err := getHostIP(host)
	if err != nil {
		return false, err
	}

	nonInterIPV4s := mustGetLocalIP4().Intersection(hostIPs)

New-Item -ItemType directory -Path $gradleInit
echo "Copy .ci/init.gradle to $gradleInit"
Copy-Item .ci/init.gradle -Destination $gradleInit

[Environment]::SetEnvironmentVariable("JAVA_HOME", $null, "Machine")
$env:PATH="C:\Users\jenkins\.java\$env:ES_BUILD_JAVA\bin\;$env:PATH"
$env:JAVA_HOME=$null
$env:SYSTEM_JAVA_HOME="C:\Users\jenkins\.java\$env:ES_RUNTIME_JAVA"
Remove-Item -Recurse -Force \tmp -ErrorAction Ignore

## TensorFlow models are programs

TensorFlow
[**models**](https://developers.google.com/machine-learning/glossary/#model) (to
use a term commonly used by machine learning practitioners) are expressed as
programs that TensorFlow executes. TensorFlow programs are encoded as
computation
[**graphs**](https://developers.google.com/machine-learning/glossary/#graph).
Since models are practically programs that TensorFlow executes, using untrusted

describes the peculiarities that apply when writing assembly code to interact with Go.
</p>

<p>
The most important thing to know about Go's assembler is that it is not a direct representation of the underlying machine.
Some of the details map precisely to the machine, but some do not.
This is because the compiler suite (see
<a href="https://9p.io/sys/doc/compiler.html">this description</a>)
needs no assembler pass in the usual pipeline.

New-Item -ItemType directory -Path $gradleInit
echo "Copy .ci/init.gradle to $gradleInit"
Copy-Item .ci/init.gradle -Destination $gradleInit

[Environment]::SetEnvironmentVariable("JAVA_HOME", $null, "Machine")
$env:PATH="C:\Users\jenkins\.java\$env:ES_BUILD_JAVA\bin\;$env:PATH"
$env:JAVA_HOME=$null
$env:SYSTEM_JAVA_HOME="C:\Users\jenkins\.java\$env:ES_RUNTIME_JAVA"
Remove-Item -Recurse -Force \tmp -ErrorAction Ignore

## `fastapi run` { #fastapi-run }

Exécuter `fastapi run` démarre FastAPI en mode production par défaut.

### Concurrency + Parallelism: Web + Machine Learning { #concurrency-parallelism-web-machine-learning }

With **FastAPI** you can take advantage of concurrency that is very common for web development (the same main attraction of NodeJS).

But you can also exploit the benefits of parallelism and multiprocessing (having multiple processes running in parallel) for **CPU bound** workloads like those in Machine Learning systems.

### Mémoire du serveur { #server-memory }

Par exemple, si votre code charge un modèle de Machine Learning de **1 Go**, lorsque vous exécutez un processus avec votre API, il consommera au moins 1 Go de RAM. Et si vous démarrez **4 processus** (4 workers), chacun consommera 1 Go de RAM. Donc au total, votre API consommera **4 Go de RAM**.

Et si votre serveur distant ou votre machine virtuelle n'a que 3 Go de RAM, essayer de charger plus de 4 Go de RAM posera problème. 🚨

## What Does Deployment Mean { #what-does-deployment-mean }

To **deploy** an application means to perform the necessary steps to make it **available to the users**.

For a **web API**, it normally involves putting it in a **remote machine**, with a **server program** that provides good performance, stability, etc, so that your **users** can **access** the application efficiently and without interruptions or problems.