Pivotal spins off Cloud Foundry, gains SAP and Rackspace support
Enterprise platform as a service provider Cloud Foundry has been spun off from Pivotal, the company said Monday. Rather than being overseen by one company the open source PaaS will be developed by the Cloud Foundry Foundation, which also recently added SAP and Rackspace to its impressive cross-section of industry supporters. But will participating vendors and cloud service providers be able to avoid some of the pitfalls inherent to other open source cloud projects and create differentiated Cloud Foundry-based offerings?
Cloud Foundry is already Apache 2.0 licensed to operate as an open source project, letting users tweak the core code to their development needs. But today’s announcement means Cloud Foundry’s development efforts will be organised in a much more formal way, following a similar path to another joint industry open source project, OpenStack.
The organisation already counts cloud and enterprise IT heavyweights like IBM, EMC, VMware, Pivotal, HP and CenturyLink (Savvis) as sponsors, who are said to be chipping in over $1m each towards the recently created foundation. But it also announced Monday that it has gained the support of German software giant SAP and, interestingly, Rackspace – which in the not-so-distant past announced its own open source platform as a service programme, Solum.
The announcement can be very much interpreted as a coming of age for the platform as a service model. Critically, the fact that these cloud heavyweights (and the nearly 800 developers working on Cloud Foundry) are backing the newly established foundation speaks to the potentially vast sums PaaS could net them.
OpenStack – the miner’s canary?
But it raises more questions than it satisfies, at least if OpenStack is in any way a guiding star here.
Rackspace’s embrace of Cloud Foundry means the company is (smartly) hedging its bets in the PaaS space; but much like OpenStack, where Rackspace has ‘bet the farm’, the noise around Cloud Foundry has, so far, been much louder than its customer adoption rates seem to merit.
It’s also fairly well known that OpenStack has created tensions among vendors, something that hasn’t seemingly let up despite the evolution of its organisation – and something which risks gripping the recently announced Cloud Foundry.
For what it’s worth, Cloud Foundry doesn’t (yet) suffer the more substantial challenges currently inherent to the OpenStack community – mainly, a lack of clarity around what OpenStack does and what it doesn’t do, and, that most participating vendors seem reticent to clarify how they plan to make money from the platform (and appropriately articulate their pitch to customers).
Several years on, even by Red Hat’s own estimations, commercial adoption of OpenStack is still minimal.
Beyond Tier 3 with Iron Foundry, there haven’t been many others to have successfully had a go at popularising a distribution of Cloud Foundry – though whether CenturyLink is making money from commercial support yet is another story entirely. Contrary to being all things to all people, what made Iron Foundry so successful is that it was designed to solve a very specific problem (putting .NET applications in the cloud), and leveraged the technology at the core of Cloud Foundry plus some secret sauce to do it.
If you build it, they will come
The Cloud Foundry announcement is also interesting for what it lacks – mainly, the inclusion of Red Hat. Unlike Rackspace, Red Hat seems content to continue focusing on OpenShift, which the company’s general manager for the cloud business unit Bryan Che told Business Cloud News is a big part of its recently revamped ‘open hybrid cloud’ strategy.
But is there really space for two (or more) open source platform as a service projects? In the world of cloud management and orchestration OpenStack seems to be garnering the lion’s share of attention from vendors and developers, much more than rival open source project CloudStack, which in November gained its first commercially supported distribution.
At the moment, CloudStack’s future seems vested in whether its main backer Citrix (which last year dropped support for OpenStack to focus on CloudStack) can keep it going or truly differentiate it, which echoes the Red Hat / OpenShift Origin story so far. For the moment, however, vendors and cloud service providers seem to have spoken: OpenStack, despite its shortcomings, is the horse to back in this race.
But the real question isn’t what will happen to the open source PaaS efforts led by Red Had and Rackspace, or how the CloudStack / OpenStack rivalry will conclude. It’s whether the vendors that have signed up to the recently formed Cloud Foundry Foundation will be able to differentiate their PaaS offerings enough to ably convince developers, as Iron Foundry did for Windows devs, without steering too far away from the technology standards at its core – that’s where the uncertainty really lies.