Developers and cloud operators from all over the world gathered in sunny Santa Clara at the end of May for Cloud Foundry Summit 2016. The event is a premier event for one of tech industry’s leading open source cloud platforms. Progressive enterprises like Allstate talked about how they adopted Cloud Foundry. There were standing room only sessions like “Cloud Foundry and Containers” by IBM’s Julian Friedman. The room was packed all the way to the back door for that one. Springer Nature’s Daniel Otte and Simon Johansson had the best slides and delivered an entertaining keynote. Credit to Simon’s wife, who has a lot of talent and toys, for making those slides.
Tech trailer blazer Paul Maritz spoke during the closing keynote. He described a pattern where technology has one open and one closed winner. For example, Android and iOS are, respectively, the open and closed winners among mobile platforms. Linux and Windows are the open and closed winners among operating systems. The question raised is, will Cloud Foundry emerge as the open winner for cloud?
There is a lot of positive momentum for Cloud Foundry based on statements made by enterprises represented at the summit. Comcast’s Tim Leong said that they now “run more than 900 apps on Cloud Foundry”. Allstate’s Doug Safford said that they have upped their spending on extreme agile development stating, “Last year we spent $5M. This year, we spent $30-50M”. The 2016 summit had participation and speakers from a number of enterprises, including Comcast, GE, GAP, Ford, Allstate, Orange, Kaiser Permanente, BNY Mellon, Visa, Thomson Reuters, Swisscom and the Government of Australia.
Relative to other PaaS alternatives, Cloud Foundry seems to have a clear lead. Adoption, however, is not easy. Allstate’s Doug Safford gave an outstanding talk about how his company adopted Cloud Foundry. The biggest challenge, he noted, has not been technology. The biggest challenge is culture. He highlighted this using a chart titled the “Frozen Middle”, showing that there is a large group in existing companies that resists change.
Beyond cultural challenges, there are also competitive alternatives. As Altoros’ Renat Khasanshyn said, “Cloud Foundry is technology, and as such, is subject to disruption.” The competitor that looms is containers. The aforementioned “Cloud Foundry and Containers” session spoke about containers and Cloud Foundry as being cooperative technologies and not competing ones.
Containers are actually an essential part of Cloud Foundry and most PaaS solutions. Cloud Foundry provides container orchestration through Warden or Garden. The difference is that a PaaS abstracts away the container technology. In my opinion, whether they compete or cooperate, has yet to play out.
On one hand, Cloud Foundry provides tremendous advantages for application developers. All a developer has to do is hand over (or push) code and the platform will do all the packaging (using appropriate buildpacks). Cloud Foundry will automatically determine requirements (e.g. which buildpack is required), install it where it needs to run, attach required services and setup all routing/proxies and get a new app up and running. With container technologies like Kubernetes and Docker Swarm, packaging becomes the developer’s responsibility. As Chris Blackburn said in his “IaaS vs. PaaS” blog, “Even if you’re under good configuration management within your IaaS fleet, PaaS is likely to deliver you a runtime environment in a fraction of the time.”
On the other hand, many IT teams see Cloud Foundry as being complex and difficult to implement. There’s also some sentiment that PaaS in general has too many parts and needs time to mature and become more stable. Containers, which one might say are a subset of PaaS, can satisfy many developer needs. Containers can also be much simpler for an IT organization. They are easier to implement and require less process and cultural change to get started. As we look to answer the question “Will Cloud Foundry emerge as the open winner for cloud?”, we should also ask another one. Is Cloud Foundry worth the (implementation and cultural change) cost?