The Rackspace Experience, Part 1: Modeling Your OpenStack Cloud

Everyone knows OpenStack has evolved over the past several years to become a leading software stack for private and hybrid clouds. As one of the largest open source endeavors, OpenStack serves as a reminder that proprietary cloud isn’t the only option — for many organizations which value leading-edge capabilities, OpenStack is a compelling choice.

And Rackspace is a leading expert on OpenStack. As one of the initial creators (along with NASA), Rackspace has been an OpenStack provider and contributor longer than anyone. We’ve run thousands of nodes for hundreds of customers, and we’re able to consistently exceed our 99.99 percent uptime guarantee for all OpenStack core services. Our OpenStack private cloud solution, Rackspace Private Cloud (RPC): Powered by OpenStack, delivers the agility and efficiency of a public cloud, along with the security, control and performance of a dedicated environment.

But many prospects wonder how we do it. How do we successfully deliver OpenStack clouds for organizations of any size? To answer those questions, we’ll use this three-part series to cover our approach to designing, deploying and operating OpenStack private clouds as a managed service. Let’s begin by taking a look at our design approach.


Rackspace Private Cloud uses a six-phase solution modeling process to help you define the parameters of your private cloud. This post will cover the first five phases, and we’ll concentrate on the sixth phase in part two.

The documentation we gather during this process ensures you won’t be asked the same question multiple times by different team members and that we provide a smooth transition from designing to deploying and operating your cloud.

Our consultative approach starts with getting a clear understanding of your objectives and discussing how RPC can help meet your needs. After our initial discussion, you have the option to have RPC solution architects meet with you onsite to conduct a two-day design session.

During this session, we assess your current state environment and discuss your business requirements, technical requirements and sizing needs for your cloud. This process includes gathering information about your applications, infrastructure environment, security and compliance needs, systems management capabilities, capacity management and operational readiness.

All of this information is documented in your “customer requirements” report and reviewed with you for accuracy. After you’ve validated your report, we put together a detailed cloud design document for your environment that includes hardware, software and network requirements, your cloud architecture and summary of architectural design decisions, a network diagram and several cloud specific details like your availability zones, tenant configuration, user roles, virtual instance design and monitoring specifications.

Thanks to our years of experience, we have strong recommendations for best practices from networking through backend storage. Our experience lends itself to a very flexible approach to leveraging your existing infrastructure if that’s your preference while implementing best architecture practice for the highest security and availability. If your preference is to have your cloud in one data center or in a hosted data center, our experience streamlines your choice and accelerates implementation.

Once you approve your cloud design, we begin preparing for implementation by creating a pre-implementation document, which summarizes the key information associated with your cloud delivery including deployment location, support services, Glance backend storage selection, networking configurations, tenant names and availability zones.

Your cloud deployment is scheduled following your approval of the pre-implementation document. We recently initiated an NPS program for our pre-implementation engagement, helping us evaluate our processes, our efficiency and our quality — helping us understand where there is room for improvement. Our surveyed customers gave us a 100% NPS score, demonstrating that our engagement urgently identifies and efficiently solves any problems for better customer outcomes.

As you can see, our process is as methodical as it could be. It’s been improved and enhanced through hundreds of successful customer engagements. Unlike some vendors who do more of an ad hoc assessment, we’ve worked out a distinctive process that accelerates time to value while minimizing impediments to success.

In the next post in this series, we’ll offer a deep dive into implementation, helping you understand how we have the flexibility to deliver an OpenStack private cloud in your data center, in our data center, or in a third-party data center without added complexity or compromise.

To learn more and ask questions about whether private cloud as a service might be a good fit for your organization, take advantage of a free strategy session with a private cloud expert — no strings attached. SIGN UP NOW.

Steve Ribble serves as Director of Operations for Rackspace’s OpenStack Private Cloud offering. In this role, Steve leads technical and operational teams delivering a world-class customer experience through open source technology stacks. Prior to his current role, Steve spent seven years assisting Rackspace’s enterprise offering build-out where he led a service delivery group for its largest and most complex customer base. His goal over the last 10 years has been a relentless focus on the end user experience taking traditional technical support structures and transforming them into a customized, value-differentiated managed services model, at scale. Steve holds a B.S. in Business Management from Trinity University in San Antonio, where he also currently resides with his beautiful wife Lyndsey and two wonderful sons, William and Caden.


  1. Thanks for the blog post.

    It could be easier to understand if abbreviations like “SFDC” be explained below the diagram.


Please enter your comment!
Please enter your name here