Overview of container and its difference from virtual machine

Container:? Containers are lightweight operating system-level virtualization that allows us to run applications and their dependencies in a resource-isolated process.

All components needed to run the application will be packaged into a mirror and can be reused. When mirroring is performed, it runs in an isolated environment and does not * * * enjoy the host's memory, CPU and disk, which ensures that the processes inside the container cannot monitor any processes outside the container. The application process in the container runs directly in the kernel of the host. There is no kernel in the container, and there is no hardware virtualization, but the process is encapsulated and isolated.

Virtual machine: it usually includes the whole operating system and its applications. Do you need to run another one? hypervisor？ To control the virtual machine.

After a set of hardware is virtualized, a complete operating system runs on it, and the required application processes run on the system.

Capacity size:

Virtual machine: the size is usually a few g,

Container: The container is light in weight and its size is within 100 meter.

Performance:

Virtual machine: It takes several minutes to start the operating system and initialize the managed application.

Container: It has better performance and can be started in almost a few seconds.

Architecture:

1. Agile environment: The biggest advantage of container technology is that it is much faster to create container instances than to create virtual machine instances, and lightweight scripts of containers can reduce the overhead in terms of performance and size.

2. Improve productivity: Containers improve the productivity of developers by eliminating cross-service dependencies and conflicts. Each container can be regarded as a different microservice, so it can be upgraded independently without worrying about synchronization.

3. Version control: Each container image has version control, so that different versions of containers can be tracked and differences between versions can be monitored.

4. The running environment is portable: the container encapsulates all the relevant details needed to run the application, such as application dependencies and operating system. This makes it more flexible to transplant images from one environment to another. For example, the same image can run in Windows or Linux, or in a development, test or staging environment.

5. Standardization: Most containers are based on open standards and can run on all mainstream Linux distributions and Microsoft platforms.

6. Security: The processes between containers are isolated from each other, and so is the infrastructure. In this way, the upgrade or change of one container will not affect other containers.

1. Increased complexity: As the number of containers and applications increases, so does the complexity. Managing so many containers in a production environment is a challenging task. Kubernetes？ And then what? Messos? Tools such as management have a certain number of containers.

2. Native Linux support: Most container technologies, such as? Docker based on Linux container (LXC), running container in Microsoft environment is a little clumsy than running container in native Linux, and daily use will also bring complexity.

3. Immature: Container technology is relatively new in the market, and it takes time to adapt to the market. The available resources among developers are limited. If developers encounter problems, it may take some time to solve them.

Operating system container: As Wikipedia said, "Operating system layer virtualization is a computer virtualization technology, which virtualizes the operating system kernel and allows multiple independent user spaces to exist instead of just one. These instances are sometimes referred to as containers, virtual engines, virtual private servers or prisons (FreeBSD prisons or chroot prisons). From the perspective of programs running in containers, these instances are like real computers. "

As mentioned above, the container * * * likes to host the kernel of the host, but provides user space isolation. We can install, configure and run applications in the container, just like on the host operating system. Similarly, resources allocated to containers are only visible to them. Just like, no virtual machine can get the resources of other virtual machines.

Operating system containers are very useful when a large number of operating systems with the same configuration need to be configured. Therefore, containers are helpful for creating templates and can be used to create containers with a similar style to another operating system.

In order to create operating system containers, we can use container technologies, such as LXC, OpenVZ, Linux VServer, BSD Jails and Solaris zones.

Application container: As Wikipedia said, "Application virtualization is a software technology that encapsulates computer programs from the underlying operating system executed by computers. A fully virtualized application, although still executed as before, will not be installed in the traditional way. The application behaves at runtime as if it directly interacts with the original operating system and all resources managed by the operating system, but it can achieve different degrees of isolation or sandbox. "

In this case, the term "virtualization" refers to encapsulated artifacts (applications), which is quite different from its meaning in hardware virtualization, in which it is related to abstract artifacts (physical hardware).

The application container is designed to package services and run as a single process, while in the operating system container, multiple services and processes can run.

Container technologies such as Docker and Rocket are examples of application containers.