2.2.2 服务器蔓延

One of the most common deployment models for microservices is to have one microservice instance deployed on one server. In a large microservices-based application, you might end up with 50 to 100 servers or containers (usually virtual) that have to be built and maintained in production alone. Even with the lower cost of running these services in the cloud, the operational complexity of having to manage and monitor these servers can be tremendous.

NOTE The flexibility of microservices has to be weighed against the cost of running all of these servers.

对于微服务，常见的一种部署模型是，在一台服务器上部署一个微服务实例。在一个大型的基于微服务的应用，你可能会用到50到100个服务器或者容器（通常是虚拟的），而这些服务器或容器必须单独在生产环境里进行构建和维护。即便在云中运行这些服务的成本很低，但依旧不得不去管理和监控这些服务器，而且这些运营复杂性是很大的。

注意 必须要在微服务的灵活性和运行这些服务器的代价之间做一个权衡。