Mirantis CEO says OpenStack will win

Amazon Web Services and its buddy Eucalyptus, Microsoft Windows Azure, Apache CloudStack and OpenStack are all racing to be stack of software organizations will use to build their Cloud-based workloads. Mirantis' CEO, Adrian Ionel, believes OpenStack has all of the earmarks of the winner. Do you agree?

Adrian Ionel, Mirantis CEO and President, stopped by to chat about how organizations are adopting cloud computing and how, in his vision, OpenStack is positioned to win over the others. Although it is impossible for me to really do justice to our rapidly moving conversation, I'll try to summarize what was touched upon.

Trends in developing Cloud-based applications

There are a number of groups seeking to define the stack of software everyone will use to develop cloud-based workloads. Amazon Web Services and its frenemy Eucalyptus Systems hope that the stack of software defined by Amazon will win. Microsoft hopes that Microsoft's Windows Azure will become the standard. Two different open source communities, Apache CloudStack and the OpenStack believe that their approach to building cloud-based applications will win.

Ionel believes that OpenStack is likely to win because of the following reasons:

  • OpenStack was developed by a community of vendors, service suppliers and end user organizations to address real world problems. It is not proprietary and not designed to address the needs of a single vendor
  • The OpenStack foundation has already attracted more than 9,500 individual members from 100 countries and 850 different organizations. It has secured more than $10 million in funding.
  • It is now easy to deploy and manage. At this point, Lonel was happy to point out that Mirantis has developed FUEL, an open source tool designed to make installation and provisioning of OpenStack easy.
  • OpenStack is open and supports many operating systems, types of virtual machine software and networks.
  • Open source projects, such as Hadoop, can already be supported in an OpenStack environment.

While Ionel's view appears based upon what his company has seen in the market, there are other viewpoints.

Snapshot Analysis

Let's step back for a moment and look at a few foundation points for this discussion. The key questions are "What is Cloud Computing?," "What are the Elements of Cloud Computing?," and "Why Cloud Computing?" Then it would be possible to address which approach is going to win and where.

What is Cloud Computing

Cloud Computing is an outgrowth of several longstanding trends and can be seen as a slightly new way to address organizations' desire to focus only on the areas in which they produce the most added value. Other functions are given to partners that can provide those functions more efficiently and at less cost.

Cloud computing, while presented as something exciting and totally new, can be seen as just the most recent evolution in computing that can be placed at the intersection of the use of industry standard X86-based systems, outsourcing and the use of increasingly virtualized environments. While cloud computing services are available based upon Mainframes and midrange computers running UNIX or a single-vendor operating system, X86-based solutions are getting most of the industry attention.

Elements of Cloud Computing

Cloud computing is the use of computing resources (hardware and software) that are delivered as a service over a network (typically the Internet). Usually, these resources are self-provisioned, paid for using a "by-the-use" utility model, easily scaled up or down as needed and typically supported in a multi-tenant environment.

Some companies wish to host some or all of their own workloads in their own data centers rather than placing proprietary data in the hands of someone else. Complying with regulations and laws is another reason some wish to be their own cloud services provider.

Why cloud computing?

Organizations who are selecting cloud computing are seeking cost reductions in the areas of staffing, software licensing, hardware, and support. Sometimes, however, they are making this selection based upon a lack of data center space, the lack of additional power or cooling in the data center, a lack of available funding for a new project or simply wanting to limit their risk by trying out something prior to investing in systems, software and staff.

Although cloud computing is often haled as the next thing in computing, it really is just a slight extension of previous industry trends.

Can their be a single winner?

Amazon and its frenemy Eucalyptus would point out that the stack of software they're supporting holds the lion's share of the market now. While this point is true at the moment, AWS is proprietary and may not support all of the operating systems, types of virtual machine software or management tools an organization might need.

Microsoft would put forward that Windows Azure will win because it is based upon Microsoft's own industry leading operating system, its choice of virtual machine software - Hyper-V, and supports Microsoft's development tools, application frameworks, databases, etc. While this approach will be very attractive to those who have standardized on Windows, it isn't all that interesting to those who have made other selections.

CloudStack and OpenStack are both open source and are developed and supported by communities of users. They both can support a broad range of operating systems, different virtual machine software products and a whole range of development tools, databases, application frameworks and applications.

While Ionel and Mirantis are convinced that OpenStack will be the winner, it seems to me that the standard is still in play. Only time will tell which view is right.