As a sales specialist dealing with companies of all sizes, I’ve seen too many organisations fail when trying to architect, deploy and operate OpenStack clouds on their own.
That’s why I’m such a strong proponent for consuming OpenStack private clouds as-a-service.
OpenStack is an API enabled technology that can be deployed to control large pools of compute, storage and networking resources across data centers, giving you the ability to deploy a public cloud-like environment in your own data center. It’s the de facto standard for private cloud today, which is why so many companies, from startups to large enterprises use OpenStack for their production workloads.
And while it’s easy to download OpenStack and start doing it yourself, beyond a small deployment, that path is fraught with danger. Operating an OpenStack cloud at scale, with enterprise-level stability, is highly complex. That complexity is one of the main reasons fully half of all organisations fail to deploy OpenStack.
And that’s why Rackspace offers a superior alternative: OpenStack private cloud as a service.
As a co-creator of OpenStack, Rackspace recognizes both the power of the platform and its complexity. That’s why we pioneered the private cloud as-a-service model, where we provide the expertise, operating your private cloud with 24x7x365 support, monitoring and maintenance.
Unlike any other provider in the industry, we have accumulated more than one billion server hours of experience managing OpenStack clouds at scale for many of the world’s largest and most respected companies. Our offerings derive from this unparalleled maturity and experience, and has established us as the number one OpenStack service provider, one reason we were named a “Leader” in an independant evaluation.
So why do DIY OpenStack deployments fail, and what does Rackspace do that enables success? After talking with hundreds of customers, sales specialists like myself see clear patterns among those who fail. Whether or not you’re considering using Rackspace, keep reading and use these challenges as a checklist for your own efforts.
Lacking the right resources
The biggest problem, and possibly the most common, is a lack of proper resources. Finding OpenStack expertise can be a problem, especially outside major metropolitan areas. That means when you can find it, that expertise often comes at a premium price. It’s also very mobile; OpenStack experts can easily find high-paying roles elsewhere. For organisations that need to build an OpenStack team, those facts can be a challenge.
To make matters more difficult, many organisations trying to migrate from traditional IT or virtualised IT don’t have the expertise in house to move to any cloud, much less building and migrating to an OpenStack cloud. Training staff takes time, money, and effort, all of which are often in short supply, especially if, like many organizations, IT budgets are flat.
Another resource issue is a lack of unallocated data center hardware to bring to the task. Unless your company is keeping a few dozen unused servers, switches, and a storage array around, you’ll have to figure out how to migrate and consolidate workloads to free up hardware for your cloud. Or worse, you’ll have to go BUY new hardware, adding procurement decisions to your punch list.
With Rackspace, you can sidestep all these issues. As one of the largest contributors to the OpenStack community, we bring all the right resources to bear on every customer project. You can leverage hardware in any of our data centers worldwide or let us help repurpose your existing infrastructure. We also bring unparalleled operational expertise, which means you don’t need to spend the time and expense of adding new staff or training existing staff, outlays that delay reaping the value of OpenStack.
Time to value
To that point, the second problem is time to value. For most organisations, unfamiliarity with OpenStack increases the time it takes them to start deriving value from their cloud. OpenStack has a learning curve, and for most organisations, setting up an OpenStack cloud isn’t automatically intuitive or streamlined. Organisations need to answer myriad questions, including:
- What release should I choose?
- What projects should I use?
- What third party add-ons (HAProxy, for example) should I consider?
- What scale? How large will my cloud be?
- What applications should I deploy or migrate?
- Which users get to access the cloud?
- What’s my timeline?
- What are my KPIs?
- How will I monitor, secure and upgrade my cloud?
The answers aren’t always obvious, and each choice has pros and cons of its own. These questions and dozens of others too often delay OpenStack implementation for do-it-yourself deployments.
Those delays also impede institutional agility, because developers need faster development cycles in order to hit business KPIs and ultimately serve customers. This is a fundamental problem that often causes cloud projects to be scrapped.
Rackspace delivers exceptional time to value. We typically can get an OpenStack cloud into production in as little as five business days. We can quickly help you answer all the questions above, with a valuable, experienced point of view. We’ll walk you through a comprehensive, well-defined experience that’s easy to understand, keeping internal communications clear and your executives in tune with the accelerated progress.
Building a new platform is inevitably expensive. Since it’s unlikely your organisation has dozens of servers and storage arrays lying idle, odds are good your cloud needs new hardware. It also needs new operational expertise — so both CapEx and OpEx spike quickly and may continue to increase as your cloud requirements evolve.
You’ll also be stuck with an uncertain ROI. Since DIY OpenStack is resource intensive and often comes with delays, establishing a quick ROI might be impossible. Your finance teams will worry about the chances of your cloud implementation becoming a money pit, with no return on investment in sight. Those pressures can impede institutional commitment and support from other organizations that you’ll need to get the job done.
With Rackspace, those worries are under control. Rackspace customers have saved an average of $258,000 annually in operating costs per 20 servers versus DIY operations. Most importantly, with Rackspace, your costs are predictable — easy to budget for. You can plan ahead for growth and have a clear understanding of when the project delivers a positive ROI.
Beyond strategic scope
Finally, managing IT infrastructure is almost never strategic to a company’s core business. Just as organisations don’t employ teams for power generation or building construction, creating teams to handle cloud implementation and management is rightly seen as waste of time and effort for many organisations.
They want to consume cloud services, not build it themselves. Most companies would rather concentrate their efforts on customer facing innovations instead of back office updates and enhancements. Maintaining a focus on revenue generating activities is the smart move for organisations that want to compete.
With Rackspace, we manage everything to enable your competitive efforts without the hassles of building a cloud. Much like public cloud, OpenStack private cloud as a service from Rackspace may not involve any of your physical infrastructure at all — we can provide all the hardware for you. Yet unlike public cloud, while we are providing the infrastructure and the software, we’re also providing management and maintenance. Your entire organization can use its customised private cloud without ever lifting a finger to build it, monitor it or maintain it.
For all the reasons above, we know that do-it-yourself OpenStack isn’t the right approach for most organisations. Rackspace pioneered the “as a service” consumption model for OpenStack, and we’ve been delivering our 99.99 percent OpenStack API uptime guarantee for years.
To learn more and find out whether OpenStack private cloud as a service or our build-operate-transfer model might be a good fit for your organisation please visit our OpenStack overview page.