## Dikkat Edilmesi Gerekenler { #caveat }

Yoğun arka plan hesaplamaları yapmanız gerekiyorsa ve bunun aynı process tarafından çalıştırılmasına şart yoksa (örneğin memory, değişkenler vb. paylaşmanız gerekmiyorsa), [Celery](https://docs.celeryq.dev) gibi daha büyük araçları kullanmak size fayda sağlayabilir.

        }
    }

    /**
     * Sets up Guice class loading mode to CHILD, if not already set.
     * Default Guice class loading mode uses a terminally deprecated JDK memory-access classes.
     */
    protected void setupGuiceClassLoading(C context) {
        if (System.getProperty("guice_custom_class_loading", "").isBlank()) {

    }
    executor.shutdown();
  }

  // In a previous implementation this would cause a stack overflow after ~2000 futures chained
  // together.  Now it should only be limited by available memory (and time)
  public void testSetFuture_stackOverflow() {
    SettableFuture<String> orig = SettableFuture.create();
    SettableFuture<String> prev = orig;
    for (int i = 0; i < 100000; i++) {

- kube-controller-manager currently needs a writable `--cert-dir` (default...

introduced significant changes to its garbage collection algorithm. As a result, we observed an increase in memory usage for kube-apiserver in larger an heavily loaded clusters up to ~25% (with the benefit of API call latencies drop by up to 10x on 99th percentiles). If the memory increase is not acceptable for you you can mitigate by setting GOGC env variable (for our tests using GOGC=63 brings memory usage back to original value, although the exact value may depend on usage patterns on your cluster)....

现在仍然是对的。

因此，为了能够同时拥有**多个进程**，必须有一个**单个进程侦听端口**，然后以某种方式将通信传输到每个工作进程。

### 每个进程的内存 { #memory-per-process }

现在，当程序将内容加载到内存中时，例如，将机器学习模型加载到变量中，或者将大文件的内容加载到变量中，所有这些都会消耗服务器的一点内存 (RAM) 。

多个进程通常**不共享任何内存**。 这意味着每个正在运行的进程都有自己的东西、变量和内存。 如果您的代码消耗了大量内存，**每个进程**将消耗等量的内存。

### 服务器内存 { #server-memory }

* Docker in Swarm Mode
* Systemd
* Supervisor
* Bir cloud provider tarafından servislerinin parçası olarak içeride yönetilmesi
* Diğerleri...

## Replikasyon - Process'ler ve Bellek { #replication-processes-and-memory }

FastAPI uygulamasında, Uvicorn'u çalıştıran `fastapi` komutu gibi bir server programı kullanırken, uygulamayı **tek bir process** içinde bir kez çalıştırmak bile aynı anda birden fazla client'a hizmet verebilir.

That "wait for something else" normally refers to <abbr title="Input and Output">I/O</abbr> operations that are relatively "slow" (compared to the speed of the processor and the RAM memory), like waiting for:

* the data from the client to be sent through the network
* the data sent by your program to be received by the client through the network

* Docker no Modo Swarm
* Systemd
* Supervisor
* Gerenciado internamente por um provedor de nuvem como parte de seus serviços
* Outros...

## Replicação - Processos e Memória { #replication-processes-and-memory }

Com um aplicativo FastAPI, usando um programa de servidor como o comando `fastapi` que executa o Uvicorn, executá-lo uma vez em **um processo** pode atender a vários clientes simultaneamente.

* Docker Compose
* Docker в режиме Swarm (Swarm Mode)
* Systemd
* Supervisor
* Обработка внутри облачного провайдера как часть его услуг
* Прочие...

## Репликация — процессы и память { #replication-processes-and-memory }

В приложении FastAPI, используя серверную программу (например, команду `fastapi`, которая запускает Uvicorn), запуск в **одном процессе** уже позволяет обслуживать нескольких клиентов одновременно.