Home > Blogs > OpenStack Blog for VMware

VMware & Rackspace Collaborate on an Interoperable OpenStack Cloud Architecture

This week at VMworld 2015 in San Francisco, VMware and Rackspace announced a new interoperable OpenStack cloud architecture. OK, that’s a big description for something that is actually very simple in concept, and highly valuable to customers who want to quickly get up and running with production-grade OpenStack clouds without a lot of the complexity associated with OpenStack implementations.

The basic premise of this architecture is that the end game for organizations is, and should be, a focus on delivering OpenStack Infrastructure as a Service to their users. If you write your applications and infrastructure automation using standard OpenStack APIs, then the automation should work with any OpenStack cloud regardless of the underlying technologies. This promise of OpenStack has enormous value for customers. If your application and infrastructure automation are portable, then businesses can move across clouds, leverage regional OpenStack clouds to expand business in new geos and/or swap out vendors to meet their business requirements.

With the interoperable OpenStack cloud architecture, customers can start with either VMware Integrated OpenStack or Rackspace Private Cloud powered by OpenStack. Customers can build all their infrastructure automation using standard OpenStack APIs on top of these platforms. Our two companies will work to ensure this automation works with the respective OpenStack clouds. This is inline with the direction the OpenStack Foundation is taking to enable multi-cloud environments via identity federation.

We are excited to announce the interoperable OpenStack cloud architecture and look forward to engaging with customers. Here’s what the architecture looks like:

VMW+RAX Interoperable_OpenStack_Architecture

For customers that want to run a heterogeneous infrastructure underneath the OpenStack API layer, we believe the best approach is to take a multi-vendor approach to their OpenStack deployment. In this way, their environments are optimized for the underlying infrastructure, which improves operations and simplifies deployment and management of their OpenStack clouds. We are collaborating with Rackspace on exactly this type of multi-vendor architecture for OpenStack – one that removes lock-in at the infrastructure layer.


2 thoughts on “VMware & Rackspace Collaborate on an Interoperable OpenStack Cloud Architecture

  1. David H Deans

    Amr, you said “Customers can build all their infrastructure automation using standard OpenStack APIs on top of these platforms.”

    Will this automation focus on initial deployment provisioning, or do you foresee automation also as a part of ongoing cloud workload management?

    1. Trevor Roberts Jr.Trevor Roberts Jr.

      Hello David,

      Automation is a key component of cloud workloads as you well know. Application developers may begin dabbling with simple automation for one-off instance provisioning either using the APIs directly or via tools like HashiCorp’s Vagrant.

      When the comfort level increases, higher-level tools like OpenStack Heat, HashiCorp’s Terraform, and/or CI/CD systems such as Jenkins, may be employed by Developers for total lifecycle management of their application infrastructure.

      Thanks for reading!


Leave a Reply

Your email address will not be published. Required fields are marked *