- Developer Tools
Phil Whelan, March 13, 2014
The Cloud Foundry project is a little older than most people realize.
It first appeared on ActiveState's radar back in 2011 when VMware announced a beta of an open-source project they called "Cloud Foundry". This was a Platform-as-a-Service (PaaS) solution that integrated with its Java development framework, Spring. It also supported Ruby and Node.js applications.
This project was mostly written in Ruby and it is similar to the current Cloud Foundry that we now know and love. But the story goes back a little further.
Thanks to the WayBackMachine, we can see cloudfoundry.com a year earlier than that (August 2008), when it was "in alpha test".
The first commit of Cloud Tools was in January 11th, 2008, but maybe it is a stretch to say that Cloud Foundry is this old.
Back To The Future
When VMware announced the open-source project back in April 2011, ActiveState was building a PaaS of its own. This was based on the requirement that ActiveState saw from its ActivePerl and ActivePython customers. There was just no good way for Enterprise IT to deploy applications in these dynamic languages.
It made sense for ActiveState to join ranks with VMware to offer Perl and Python implementations for the Cloud Foundry open-source project. This was another step towards making Cloud Foundry more of a polyglot solution. Python was adopted in the project and ActiveState moved forward with its own Enterprise PaaS solution, now based on Cloud Foundry.
The first production release of Stackato came out on February 29, 2012 and ActiveState quickly acquired its first few customers. ActiveState became experts in the code-base, building on-top of it, fixing bugs and generally hardening it into the Stackato solution.
As the adoption of the project grew, we saw other commercial solutions appear, but generally these were of "public PaaS" variety. The first in this space was AppFog. They released their public PaaS built on-top of Cloud Foundry. HP Cloud later released a beta of their public PaaS, which is built on-top of Stackato.
Due to the decoupled nature of the Cloud Foundry components, it was easy for others to join the now growing ecosystem and build complimentary plug-in components. This is what Iron Foundry did with their .NET solution for Cloud Foundry. This was a DEA (Droplet Execution Agent) that worked with either the base Cloud Foundry or Stackato.
Later we would see another similar Windows-based .NET solution from Uhuru. Recently, Uhuru announced that they are also open-sourcing this technology.
While VMware did continue to develop Cloud Foundry throughout 2012, they did not deliver a commercial solution.
As 2012 came to an end, it was clear that development in the Cloud Foundry project was slowing and it felt like Stackato was running ahead with features. While this was good for us, it was not good for the ecosystem. Up to this point, the Cloud Foundry project had not changed very much and VMware still kept a tight grip on their open-source project.
In December 2012 VMware announced that they would spin out the Cloud Foundry project and other assets into Pivotal. Pivotal received significant investment from GE. Paul Maritz, who had recently been CEO of VMware, was to be CEO of Pivotal and drive the project forward. This hand-over of the project from VMware to Pivotal occurred in April of 2013, and so began the next generation of Cloud Foundry.
Previously, the Cloud Foundry project did not have an official version. Out of the gate, Pivotal began major rewrites of the code-base and components. They labeled them with "next generation". This quickly became what we now know of as Cloud Foundry "v2" and everything that went before was retroactively labeled "v1".
The code-base was built much more cleanly, with good code re-use, a proper and consistent REST interface, and an overhaul on both internal and external APIs. A definite fresh start.
In September of 2013 we saw the community come together in-person for the first time at the Cloud Foundry focused conference, "Platform". It seemed that everyone involved in the community, up to that point, came out. Our friends from NTT and Rakuten traveled from Japan to San Francisco.
It was here that we first saw signs of IBM's involvement in the Cloud Foundry project, with their "Blue Docs" project. This was an internal project led by University of British Columbia student, Nima Kaviani. This was a way for IBM to understand the many moving parts of Cloud Foundry better. It cataloged the message bus (NATS) messages and HTTP communication across a Cloud Foundry cluster. This gave IBM a blue-print of Cloud Foundry's inner workings and allowed their development team to more quickly get up to speed.
End Of The Line
There was some vocal dismay from the non-vendor side of the community, who had invested heavily in the first version of Cloud Foundry, that there was no route from v1 to v2. While ActiveState's Stackato did provide a migration for users of its CFv1-based release to the newer CFv2-based release of Stackato, it was unfortunate that there was no official support from Cloud Foundry for other users of the project.
In July of 2013, IBM and Pivotal announced they intended to work with the community to establish an open governance model.
In August of that year, the Cloud Foundry Advisory Board was announced, of which Savvis was quickly made a member.
Members of the Advisory Board now include IBM, CenturyLink (Savvis), Piston, Intel, Pivotal, ActiveState, Stark & Wayne, Canonical and CloudCredo.
Towards the end of 2013, Cloud Foundry version 2 was beginning to take shape and both ActiveState and Pivotal were working on commercial solutions built on this new version of Cloud Foundry. Both announced major releases close to the end of the year, setting the stage for 2014 to be a very interesting year.
As we entered 2014, the dust had settled on the transition from VMware to Pivotal and from version 1 to version 2. It was clear that the project was moving forward on a much better footing.
While still not perfect as a non-biased ecosystem, it was a vast improvement from what had come before. More importantly, it was moving in a positive direction. The time for pull-requests to be processed was down from 3 months to days as Pivotal aligned its Agile methodologies with the nature of a sizable open-source project. Things were now happening more out in the open and the Advisory Board was a promise of better things to come.
So far, 2014 has not disappointed.
January and February saw the first 2 Advisory Board meetings. These were open to anyone to listen in. It was also possible to participate via the text chat room. Both meetings were productive and extremely informative.
Days prior to the February meeting, it was announced that an independent foundation was to be formed to help govern the future direction and development of the Cloud Foundry project. This was great news for ActiveState, great news for the ecosystem of Cloud Foundry, and great news for PaaS in general.
So far, the platinum sponsors of the foundation are EMC, IBM, HP, Pivotal, Rackspace, SAP and VMware. Gold sponsors include ActiveState and CenturyLink.
A foundation helps level the playing field. Yes, this has brought more heavy-weight competition into the PaaS space, but this is exactly what this new and growing market needs. It also gives ActiveState the breathing room to be more actively involved in the open-source project, as it is now decoupled from a single commercial entity.
We are looking forward to the Cloud Foundry Summit taking place in San Francisco in the summer, the formation of the foundation, and a more open governance structure for the future of Cloud Foundry.
Update: To be clear, the Spring version of Cloud Foundry and Ruby version of Cloud Foundry, that came later, were different products and only related by the fact that they were both PaaS solutions, both shared the same name, both owned by VMware and both hosted on cloudfoundry.com.
Share this post:
The PaaS landscape is changing. Many of the anticipated shifts are yet to come. Most of the expected changes are based on predictions of what the big players should do to align themselves for a prosperous future.
Last week, the nonprofit foundation tasked to oversee the open source Cloud Foundry project was announced. Like large stones being placed on one side of a weighing scale, things tipped in favor of Cloud Foundry as the open-source PaaS solution to bet on.
Earlier this week there was an announcement about the Cloud Foundry foundation, which Adrian Coyle from Pivotal referred to as "the worst kept secret."
"We have been working on how we can open up Cloud Foundry as a foundation and put it in a place to be owned and developed as a shared industry asset. The hope is to have the foundation officially launched later this year."
Work is going into drafting bylaws for which there is a first draft. There is a lot of work to incorporate the foundation. Gathering feedback from the group is of paramount importance.
Chris Ferris from IBM said, "There have been some great posts and coverage [on the foundation news]. Everyone is very excited. It's an important next step for Cloud Foundry. It is an important project for Enterprise PaaS."
Subscribe to ActiveState Blogs by Email
Share this post: