Open source software for creating private and public clouds

OpenStack software controls large pools of compute, storage, and networking resources throughout a datacenter, managed through a dashboard or via the OpenStack API. OpenStack works with popular enterprise and open source technologies making it ideal for heterogeneous infrastructure.

Hundreds of the world’s largest brands rely on OpenStack to run their businesses every day, reducing costs and helping them move faster. OpenStack has a strong ecosystem, and users seeking commercial support can choose from different OpenStack-powered products and services in the Marketplace.

The software is built by a thriving community of developers, in collaboration with users, and is designed in the open at our Summits.

The open infrastructure landscape is evolving. Users integrate dozens of open source tools into a modern stack reaching beyond the scope of OpenStack, so we re-organized the Summit to focus on specific problem domains. 

WHAT IS OPENSTACK?

OpenStack is a cloud operating system that controls large pools of compute, storage, and networking resources throughout a datacenter, all managed and provisioned through APIs with common authentication mechanisms.

A dashboard is also available, giving administrators control while empowering their users to provision resources through a web interface.

Beyond standard infrastructure-as-a-service functionality, additional components provide orchestration, fault management and service management amongst other services to ensure high availability of user applications.

Vision for OpenStack Clouds

Purpose

This is a living document. Its purpose is to document the OpenStack community’s vision for the output of the OpenStack project as a whole, as it evolves over time. It is also aspirational, not descriptive, in nature. That is, it describes the OpenStack that the community is committed to working toward, not the OpenStack that has existed at any particular point in time.

Project teams can refer to this document when evaluating proposed features and designing interfaces, to help ensure that their designs fit comfortably within a larger structure and contribute to the overall landscape of application deployment patterns.

Amongst other things, the Technical Committee will refer to this document when assessing new project applications to determine whether they fit with the overall technical direction of OpenStack. If you are working on a project that does not appear to fit with the vision as described here, that does not necessarily mean that your project can never become an official OpenStack project, but it may mean that you should submit a patch to this document to further flesh out the vision. If this is the case, you should do so as soon as possible – there is no need to wait until you are ready to submit a new project application.

The Technical Committee may also use this document to evaluate the suitability and priority of proposed project-wide goals.

Scope

The scope of this document is limited to the cloud services that an end-user interacts with. This corresponds to the main ‘OpenStack’ bucket and parts of the ‘OpenStack Operations’ bucket in the OpenStack project map. While OpenStack also has other kinds of official projects (e.g. deployment tools and client libraries), nothing at all can be inferred about our vision for them from this document.

The Pillars of Cloud

There are at least as many different opinions of what ‘cloud’ means as there are software developers. However, we can all agree that cloud does mean something. Cloud computing promotes more efficient utilization of resources by reducing the transaction costs involved in provisioning and deprovisioning infrastructure to near zero, and it is able to do so because it differs in qualitative ways from previous models of computing (including virtualization). We identify two in particular.

These concepts are not always applicable to all aspects of the system, but we expect all services in OpenStack to conform to them wherever they are applicable, either directly or by working in conjunction with other services.

Self-service

Clouds are self-service. They provide users with the ability to deploy applications on demand without having to wait for human action or review in the loop. The cloud has no ticket trackers. This requirement has a number of immediate consequences.

First, cloud services must provide robust multi-tenancy. In order to securely serve multiple users (or groups of users) without any human review, resources must be isolated between tenants of the system so that the resources controlled by one tenant have neither access to nor impact upon resources controlled by other tenants.

Cloud services must also have some mechanism to ensure that capacity is only utilized when the value to the user of doing so exceeds the opportunity cost to the operator of providing it. In public clouds this is typically accomplished by charging users for the resources consumed. Quotas are used to limit risk for both users and cloud operators. For users, the risk is of an unexpectedly-high bill, particularly when the level of resource consumption is partially under automatic control. For operators it is reaching a level of utilization where opportunity costs increase non-linearly. Private clouds will often require the same sorts of monitoring and reporting capabilities used for billing, even if they do not make use of a chargeback mechanism and rely on quotas alone as the sole technical measure to control resource consumption.

Project team reflections on this vision

When this vision was published, project teams were encouraged to write a self-evaluation or reflection to determine how their project compares to this vision. A compilation of these self-evaluations is below.

  • Keystone
  • Placement
  • Searchlight
  • Zun