知识卡片
容器共享内核并隔离进程
容器通过操作系统机制隔离进程视图与资源,通常共享主机内核,区别于各自运行来宾内核的传统虚拟机。它可把应用依赖封装成可分发镜像,便于一致部署;隔离程度和资源边界仍取决于配置,不能把它当作绝对独立的另一台机器。
这张卡片看不懂?
这张卡片已通过人工审查,不再接受失败反馈。
已有反馈 · 0 条
还没有人提交反馈。
在整书笔记中的位置
原书出处 · 1 处
- Building a Cloud Native Service · 查看原书依据
containers are not virtual machines20: they don’t use hypervisors, and they share the host’s kernel rather than carrying their own guest operating system. Instead, their isolation is provided by a clever application of several Linux kernel features, including chroot, cgroups, and kernel namespaces. In fact, it can be reasonably argued that containers are nothing more
参考来源
- 位置:《Cloud Native Go》第05章《Building a Cloud Native Service》(来源项目book-sources;源文件:extracted/Cloud Native Go (Matthew A. Titmus)/_epub-src/OEBPS/ch05.xhtml)
- 结论依据:容器通过操作系统机制隔离进程视图与资源,通常共享主机内核,区别于各自运行来宾内核的传统虚拟机。原文相应定义、示例或限制为此结论提供依据;补充纠错与版本边界在正文就近标明。
- 原始内容:containers are not virtual machines20: they don’t use hypervisors, and they share the host’s kernel rather than carrying their own guest operating system. Instead, their isolation is provided by a clever application of several Linux kernel features, including chroot, cgroups, and kernel namespaces. In fact, it can be reasonably argued that containers are nothing more