A brief introduction to the OpenStack Magnum project

Source: Internet
Author: User

Background

The Magnum project, a young project that increased OpenStack in November 2014, was initiated by Rackspace, which is positioned to provide the API framework for container-as-a-service (Container as a services), which is planned for October 2015 version number is mature.

We know that the Nova project in OpenStack now supports the Docker container (the container as a virtual machine tube) in the form of Nova-docker. But in the actual use, will find a lot of problems. After all, the Nova design is designed to manage virtual machines, and the container is significantly different from the virtual machine's behavior and characteristics. Either management or the underlying virtualization support layer is completely different.

Also, there are quite a few places where Nova supports a variety of container mechanisms (Docker, OpenVZ, Rocket, LXC, etc.) to make changes that may conflict with existing frameworks.

In addition, the Heat project also supports the official Docker plugin. To manage containers directly through Docker's REST API, and to support the advanced features of the container.

However. Resource scheduling and network functionality are not supported.

The community receives the Magnum project. On the one hand, container technology is now really hot, on the one hand, but also to a higher level of development to provide better support.

Design principle

Magnum in the design, want to invoke other container management platform API to implement the function, itself as a set of API framework, now support Docker, Kubernetes, Swarm and so on. The main advantages include multi-tenancy, multi-backend framework, intact container functions, support resource scheduling, and so on.

Assuming that Nova is a set of API frameworks that support different Hypervisor (KVM, VMWare, and other virtual machine platforms), then Magnum is an API framework that supports different container mechanisms.

Basic concepts

Order from small to large:

    • Node: The nodes that the container executes can be bare metal, virtual machines, or even containers.
    • Bay: A set of Node collections (the underlying same drive mechanism), which is the basic unit of container dispatch in Magnum. Bay is isolated between tenants.
    • Baymodle: Similar to the flavor in Nova. Defines the specifications for a Docker cluster.

The following are some of the concepts from Kubernetes.

    • Container: Container.
    • Pod: The minimal snap-in. One or more interrelated containers (typically performing the same application), executed on the same minion Node. Share the same data mount and network space, representing an instance of an application.

    • Service: consists of one or more pods that represent an abstract application service. Externally presented as the same access interface. This access can be routed through the service. Without having to know the address of the Pods in detail.

    • Replicationcontroller: Specifies the number of replicas for the pod, and RC ensures that the number of copies exists and executes.

Main services

The main services are two. Magnum API and Magnum conductor.

The former provides the calling interface to receive the python-magnumclient request. Can execute one or more instances at the same time. These requests are finally thrown to the AMQP message queue. Sent to the Magnum-conductor service.

The latter executes on the control node, which is responsible for forwarding the client's request to the detailed backend mechanism (Kubernetes API or Docker API). There is only one instance of this limitation.

Reprint Please specify: http://blog.csdn.net/yeasy/article/details/46331679

A brief introduction to the OpenStack Magnum project

Contact Us

The content source of this page is from Internet, which doesn't represent Alibaba Cloud's opinion; products and services mentioned on that page don't have any relationship with Alibaba Cloud. If the content of the page makes you feel confusing, please write us an email, we will handle the problem within 5 days after receiving your email.

If you find any instances of plagiarism from the community, please send an email to: info-contact@alibabacloud.com and provide relevant evidence. A staff member will contact you within 5 working days.

A Free Trial That Lets You Build Big!

Start building with 50+ products and up to 12 months usage for Elastic Compute Service

  • Sales Support

    1 on 1 presale consultation

  • After-Sales Support

    24/7 Technical Support 6 Free Tickets per Quarter Faster Response

  • Alibaba Cloud offers highly flexible support services tailored to meet your exact needs.