IBM today committed to contributing to Cloud Foundry, the open source platform as a service (PaaS) project originally developed by VMware and now part of Pivotal, which is a spinout by the virtualization company and EMC.
IBM officials say they hope to expand integration between Cloud Foundry and OpenStack and bring more open source elements to the project, including code contributions from organizations outside VMware and Pivotal. MORE OPENSTACK: Happy Birthday OpenStack! Now change
Cloud Foundry is an open source PaaS made up of two major components – a runtime layer for deploying and scaling applications to give them cloud-like features, such as multi-tenancy and the ability to run in virtualized environments; and an operations layer, which integrates the PaaS with underlying hardware, which can be a virtualized IaaS or bare metal. When VMware and EMC created Pivotal as a spin-out earlier this year, the Cloud Foundry product and open source project melded into that initiative.
Watters says today’s news is making public contributions IBM has been making to Cloud Foundry for months.
As part of today’s news the Cloud Foundry name will now only apply to the open source project, not the PaaS offering that VMware had. Instead, Pivotal plans to launch a Cloud Foundry distribution later this year under a new name. Pivotal also already has a Hadoop distribution and is working on other big data management products.
IBM committing to Cloud Foundry could be the first steps toward integrating the PaaS project with the OpenStack IaaS. PaaS and IaaS can be complementary with the PaaS being an application development platform and IaaS supplying the hardware resources to do so. IBM is already running an integrated Cloud Foundry-OpenStack deployment in a lab setting, says IBM distinguished engineer Christopher Ferris. A goal for Cloud Foundry, he says, is for it to be an open platform that gets contributions from a number of organizations and can run on a variety of hardware options, including not just those from OpenStack, but VMware, Amazon and others.
Another significant portion of the news is that IBM and Pivotal have worked to ensure compatibility of IBM’s WebSphere middleware apps with Cloud Foundry, meaning that any applications built to WebSphere standards will now work in a Cloud Foundry environment, giving applications cloud-like capabilities of multi-tenancy and the ability to run in virtualized environments.
Reaction to the news is not one of huge surprise by close watchers of the PaaS industry. John Treadway, senior vice president at consultant Cloud Technology Partners, which works with customers to implement PaaS deployments, says it’s a good move, but he’d like to see more partners contributing to Cloud Foundry before it really feels like an open source project and not just one being pushed by Pivotal and VMware.
Meanwhile, the integrations between Cloud Foundry and OpenStack are not unique to Pivotal and IBM; Red Hat is in a position to offer an integration of its OpenShift Origin open source PaaS with its OpenStack distribution. Treadway says end users may be more comfortable working in Red Hat or VMware environments, which could influence buying decisions.
Another factor is that Pivotal is taking a “cornucopia” approach by bundling PaaS, Hadoop and other initiatives into its services. Treadway says that may be beneficial to VMware or EMC shops, but for others it could be a little over-the-top. “Having a Swiss Army knife is great, but sometimes you need a fileting knife to really get the job done right,” he says. Other private PaaS offerings like those from Apprenda or Stackato from ActiveState offer a pure-play PaaS without all the bells and whistles that would come with an offering from Pivotal, he says.
Network World senior writer Brandon Butler covers cloud computing and social collaboration. He can be reached at BButler@nww.com and found on Twitter at @BButlerNWW.