服务网格是组织工具，而非技术工具
Original title: Service meshes are organization tools, not technical ones
The title and summary in the selected language are awaiting translation.
服务网格的价值在于以难以绕开的方式注入策略，包括连接池与 keep alive、重试、超时、数据本地性约束和 AZ 本地路由等。产品工程追求快速迭代与可靠性之间存在一定冲突，而服务网格难以被绕过，能确保默认情况下做正确的事。这也是部署服务网格最简洁的理由：它实现了职责分离，又不需要产品团队额外记住步骤。
A service mesh is great for injecting policy in a way that makes it difficult for folks to do the wrong thing: connection pooling and keep alive, retries, timeouts, enforcing data locality constraints, AZ-local routing, etc, etc.
To a certain extent, product engineering wanting to move quickly is at odds with reliability and a service mesh is difficult to subvert. Services meshes make sure the right thing happens by default.
This is probably the most succinct reason for deploying one. It allows separation of responsibilities in ways that don’t add extra steps product teams have to remember to do.
Source: Ryan Lopopolo · hyperbo.la