Another possibility is to try as much as possible to push your I/O and flow control primitives to the edges of the program or library, providing integration points for multiple backends.

In terms of receiving events, events can either be returned to the calling code immediately whenever bytes are provided, or they can be lazily produced in response to the calling code’s request.

it is usually important to preserve the datagram boundaries.

When it comes to network protocols, at a fundamental level they all consume and produce byte sequences.

This means that the core of the protocol implementation is divorced entirely from the way I/O is done or the way the API is designed.

because you have no mocking and no actual I/O

there are far fewer possible entry points and locations in the state space.

It is no secret that network I/O is particularly prone to a wide variety of unexpected failure modes that can occur at almost any time, even in the simplest cases.

服务网格利用容器之间的网络设置来控制或改变应用程序中不同组件之间的交互。

ReplicationController

把一个 service 看成是一个 pod 的逻辑分组，它提供了一个单一的 IP 地址和 DNS 名称，你可以通过它访问服务内的所有 pod。

与容器相关的一个重要概念是微服务。

Kubernetes 运行在节点 (node) 上，节点是集群中的单个机器。

最重要的是你需要认识到 Kubernetes 利用了 “期望状态” 原则。

你可以把容器看成是整个应用堆栈中的一层，每层都依赖于下层的单元。