Cloud Foundry and OpenStack and why Red Hat will not join the Cloud Foundry foundation
A few days ago, Pivotal announced that it will transition Cloud Foundry into its own Open Source foundation. Platinum sponsorship will run $1.5M over three years. Sound familiar? Clearly, the Pivotal folks are the students of recent history and are modeling their decision after Rackspace’s move to release control over OpenStack.
Now that such important backers of OpenStack as IBM, HP and Rackspace have joined the CF foundation, does it mean that the OpenStack community is content for OpenStack to stay just the IaaS and not make the attempt to move up the stack? Will we live happily ever after, as prescribed by the peace-loving piece by Jesse Proudman?
In his interview with Ben Kepes, OpenStack Foundation director Josh McKenty made a controversial prediction that OpenStack will abandon its own PaaS efforts and that Red Hat too will eventually join the CF foundation.
Quite a leap of faith, I would say, as it will be extremely hard for Red Hat to abandon its current messaging around OpenShift and its elegant attempt to morph it into OpenStack via project Solum. For many years now, Red Hat has been promoting OpenShift to its customers, and admitting that it was a mistake and betting on a different horse would be a strong blow to the company’s credibility as its customers’ most reliable adviser in the ever-changing world of open source software.
Moreover, as the industry transitions to cloud computing, Red Hat finds itself in the middle of an existential transformation. OpenStack is the only vehicle they have to morph themselves into the Cloud Infrastructure company. Red Hat needs OpenStack to protect its legacy RHEL revenues, and similarly it needs to control the PaaS layer to protect its margin in the OpenStack revenue stream. This reason alone is enough for Red Hat to continue its strong push behind project Solum.
When OpenStack itself was launched, the software was young and didn’t work that well; its competitor CloudStack was a real, functioning thing (we know because we worked with both back in the day). But it didn’t matter. What mattered was that the community momentum of the former was much stronger than that of the latter. Solum today is young. But it is also one of the most active OpenStack projects; and just like OpenStack caught up to CloudStack, Solum, Heat, Savanna, Murano and other PaaS-level projects will gradually catch up to and battle Cloud Foundry as a native PaaS on top of OpenStack.
Since both the OpenStack and CF foundations are driven by the commercial interests of the sponsoring companies, as I expressed in my earlier blogs, OpenStack will continue its ascent up the stack. And just as its member companies are not always the best of friends, there will be a constant love-hate relationship between the two ecosystems.
I agree with the assessment that Ben Kepes made in his blog:
“The major loser in this announcement would appear to be Red Hat who has both thrown its hat in with its own open source PaaS play, OpenShift, and was also a party to the Solum initiative.”
But my conclusion is different: Red Hat and OpenStack will double down on its own PaaS initiatives inside OpenStack, giving customers more interesting options to choose from.