The Enterprise IoT platform is beyond crowded these days. Telcos, enterprise software vendors, device manufacturers, and IoT startups are some of the groups trying to dominate the enterprise IoT platform space. From all these groups, the platform as a service (PaaS) providers seem to be gaining a lot of momentum among enterprise customers. Solutions like AWS IoT Hub, Azure IoT Suite or IBM IoT Foundation provide an extremely robust set of IoT capabilities for most enterprises. However, those platforms still have two major limitations in terms of support for on-premise environments and portability across different platforms.
Unlike many other infrastructure areas, the IoT space still hasn’t had the emergence of an open platform model that can be adapted across different infrastructures. In the broader PaaS space, Cloud Foundry has become the most popular open PaaS model supported in both cloud and on-premise environments. With IoT services becoming a first class citizen in PaaS solutions, we believe the enterprise IoT space can greatly benefit from a Cloud Foundry for IoT model.
Cloud Foundry for IoT can be a catalyst in the enterprise
The value proposition of a Cloud Foundry for IoT model may seem trivial. It’s not hard to imagine how this type of architecture will help to improve interoperability, openness, and portability of IoT solutions. Even though the Cloud Foundry for IoT model appears highly applicable in the consumer market, we believe there are some very specific capabilities that will make it very popular in the enterprise.
Enterprise IoT solutions are notorious for requiring significant levels of customization. An open source distribution of Cloud Foundry for IoT will help enterprises to extend or modify the platform with its proprietary capabilities. Additionally, an open source model will allow enterprises or device manufacturers to contribute to the platform, increasing its relevance in specific domains.
Many enterprise IoT solution require on-premise or hybrid deployments. The platform agnostic nature of Cloud Foundry will facilitate its viability in scenarios that are currently prohibited for cloud IoT platforms.
Complementary to the open source model, the highly extensible nature of a Cloud Foundry for IoT solution will facilitate its adaptability to complex enterprise scenarios. The lack of extensibility still remains a major challenge for most enterprise platforms.
A complete PaaS at your disposal
Similar to the cloud IoT platforms, an IoT solution based on Cloud Foundry can benefit from dozens of cloud services already existing in Cloud Foundry. This model highly contrasts with the existing generation of IoT on-premise platforms that solely focuse on IoT specific capabilities.
Imagining Cloud Foundry for IoT
Defining a Cloud Foundry for IoT solution can be as simple as incorporating IoT native services into the Cloud Foundry infrastructure. Drawing a parallel with the first generation of IoT platforms, we can identify a first group of capabilities that could be relevant to a Cloud Foundry for IoT model:
- Protocols: Support for IoT specific protocols such as MQTT, XMPP or CoApp will be an essential element to improve Cloud Foundry interoperability with different smart devices and other IoT solutions.
- Device gateway: In order to be relevant in industrial IoT settings, a Cloud Foundry for IoT model should provide capabilities to register, manage, and monitor a large number of IoT devices.
- Messaging: Even though Cloud Foundry already includes support for sophisticated messaging technologies such as NATS or RabbitMQ, IoT solutions will be required to extend those capabilities to support large throughput of messages in a scalable manner.
- Stream analytics: Adding stream analytic capabilities to Cloud Foundry will facilitate the executing of queries over real time streams of data generated by IoT devices in an enterprise topology.
- Security: Message encryption, signing, as well as access control and authentication are key capabilities of enterprise IoT solutions. Even though Cloud Foundry already provides sophisticated security capabilities, those should be extended for IoT environments.
- Device SDKs: Providing Cloud Foundry SDKs for different device toolkits such as Raspberry Pi or Arduino will be key to improve the adoption of the platform within core IoT developer communities as well as device manufacturers.
There are other capabilities that can be relevant to extend Cloud Foundry capabilities into the IoT space. However, the aforementioned group of features will make Cloud Foudry for IoT very competitive with the existing offerings from companies like Amazon, Microsoft, IBM, and Salesforce. Combined with the existing capabilities of Cloud Foundry, a new set of IoT services has the opportunity to make Cloud Foundry one of the premier IoT offerings in the market. Leveraging Cloud Foundry for IoT solutions is far from being just a theoretical exercise and companies like GE are already doing a lot of interesting work in the space.
It’s already happening: GE Predix
GE Predix is a PaaS model for the industrial enterprise. Based on Cloud Foundry, GE Predix adds a new group of services and capabilities relevant to IoT industrial solutions. Specifically, GE Predix extends Cloud Foundry with IoT specific services such as asset management, device security, real time analytics, and other capabilities relevant in IoT. By many metrics, GE Predix can be considered the most successful IoT platform in the market claiming $6B in revenue last year. In any case, GE Predix is a very strong validation of Cloud Foundry as one of the platforms that can become pivotal in the next generation of IoT solutions in the enterprise.
This article is published as part of the IDG Contributor Network. Want to Join?