Hot attention: OpenStack will become the mainstream open source cloud platform?

A recent interview with Eucalyptus CEO Marten Mickos has allowed me to rethink the question of who eucalyptus and OpenStack will become the mainstream Open-source cloud platform.

These days, the enthusiasm of the manufacturers for OpenStack is soaring. The open source project, controlled by Rackspace, is now joined by Dell, Citrix and HP. Despite the demands of these large manufacturers, Rackspace has given control of the project to the OpenStack Foundation, but as Mickos said, OpenStack is still in the research and development phase and has not yet entered the production phase.

Like many people, I've always thought that a huge openstack community would give OpenStack a key majority support, making it a leading Open-source cloud platform. But now, I'm beginning to be skeptical.

To explain my attitude change, let's review the two leading Open-source projects: Linux and Apache HTTP servers. The "Linux project" I refer to here is in its broadest sense, including the Linux kernel and all the different Linux distributions.

History tells us that when an open source project treats the value layer of the software stack, the project is always controlled by a manufacturer who uses the project directly to make money. Although there will be multiple implementations or different distributions from this project, the single vendor will be the main provider in this area. In the case of Linux, Red Hat and its Red Hat Enterprise Linux products. MySQL also fits this pattern for open source databases.

History also tells us that when an open source project treats the application layer of the software stack, the project is always controlled by a foundation. On these occasions, the project will often become part of a higher value product and thus make a profit. In short, the Open-source project itself can only indirectly make money through higher-value products. This can also be cited in two examples, one is the Apache HTTP server, used in most commercial application server products, and the other is eclipse, used in many commercial application development products.

If history repeats itself over and over again, we need to consider whether an open source cloud platform has its own value layer that can make money directly from the software stack. If it is, then a single vendor-controlled open source project will be more likely to succeed than a project controlled by the foundation.

The Open Source Foundation is really great and plays a valuable role in all kinds of open source projects. However, the different motivations of the Baixian vendors make it a dilemma for the foundation to meet the needs of the project and the interests of its manufacturers.

What the OpenStack community is looking for is a software stack that cannot be directly monetization, as Dell, Hewlett-Packard and Rackspace vendors see. Instead of using OpenStack directly to make money, these firms build a higher-value product through it. For example, Dell and HP would prefer to sell "cloud platform ready" hardware systems rather than OpenStack software products themselves. In fact, Dell has already done so.

So eucalyptus does not agree that a cloud platform is managed by a rival foundation, and the reason is clear, and its CEO, Mickos, says the company's customers are growing at an astonishing rate. Eucalyptus's staff surged from 15 to 70 last year, and a new headquarters in London to cope with the growth of markets in Europe, the Middle East and Africa. This growth just shows that eucalyptus can be directly monetization.

Ultimately, it buyers have to decide who is the right bet, Eucalyptus and OpenStack. I tend to agree that Eucalyptus is a truly valuable cloud platform, so it is possible to make money directly from the value layer of the software stack. If so, then eucalyptus should win, and OpenStack will eventually lose because of the rivalry between the members.

(Responsible editor: Liu Fen)