By Patrick McFadin, VP Developer Relations, DataStax

Data, cloud and microservices - what will change in 2020?

In 2020, cloud computing will continue to grow as the dominant approach for new application deployments. As part of this, open source software will be essential components to run those applications - from containers and orchestration management tools through to the databases, streaming services and analytics tools that will run on those cloud platforms at scale. However, the open source movement will have to change as well, from new deployment models through to new team models and approaches.

What will this emphasize in 2020? How will the open source approach have to change around cloud, and how can communities support their development?

Prediction #1 - Cloud and open source licensing issues will continue

In 2020, the open source sector will continue to be affected by the growth of cloud providers offering managed services and by open source companies launching their own approaches too. From a relationship status point of view, it will be complicated.

Over the past twelve months, a range of companies including Confluent, Elastic, and MongoDB have made changes to how they license their open source projects. The aim was to make sure that any company commercially benefitting from the open source project paid back into the community. However, this approach was not seen as ideal by many in the community. Whichever side you agree with, the impact on the world of open source as a whole will be felt next year.

Finding the right approach around open source development and cloud will continue to be an issue for open source communities, commercial open source companies and cloud providers alike. For some projects, the availability of cloud services will help to grow the community. Other projects will benefit directly as they suit the cloud deployment model or because developers find the value is worth the headache of running their own instances. However, the risk is that stricter licenses will stall both the communities and the commercial companies behind standalone open source projects alike.

Prediction #2 - Microservices will drive more open source adoption

In 2020, more companies will adopt microservices for how they design and build their apps. This adoption pattern grew in 2019, and will continue to advance next year.

The reason for this is that microservices are easier to manage and update, while it is also easier to adopt or expand more services over time. As each application component is specialised, it should lead to more adoption of open source projects that are well designed, have easy APIs to work with and meet those goals. The problem in 2020 will be using the data that these applications produce.

Microservices apps are designed to be easy to scale up, so any infrastructure will have to cope with this scalability as well. This will require database infrastructure that can ingest and manage data in the right way. Supporting these microservices-based applications effectively will lead to more open source database projects in 2020.

Prediction #3 - The role of companies in open source community development will be more obvious

Supporting open source projects relies on two things - the software itself, and the community that develops around that use case. For projects, both good software and a healthy community will be needed to grow.

In 2020, the sheer number of software projects that are available is huge. The ubiquity of the open source model means that more projects are trying to grow. This makes the community support side all the more important. Bolstering and growing the community is just as important over time, and so the role of businesses linked to open source projects will be essential in making this successful.

For projects like Apache Kafka, Apache Cassandra^TM, and Linux, the communities that developed around them were necessary for helping more companies adopt the project and expand the number of users and contributors alike. Behind these communities, commercial open source organisations were necessary to guide the early stages of development and awareness.

Today, the Cassandra project chair group includes representatives from Netflix, Instagram, Apple, and Uber. These companies have massive implementations that are reliant on Cassandra, so they lead how the project will develop. The Cassandra Enhancement Process (CEP) has been put in place to formalise decisions around the addition of major new features. The intended result is better communication and collaboration between everyone in the community. As this is led by users - rather than a single central commercial organisation - it should mean that the direction of the project meets their needs.

However, commercial organisations still have a vital role to play around open source projects like Cassandra, providing focus and support around areas that can support newer adopters. For example, producing documentation and guides on how to adopt and use a project can be just as valuable as code commits over time. 2020 will be about balancing the needs of the community alongside the interests of commercial companies, so that everyone benefits from the work that takes place. The value for the community from commercial companies is from serving that community and ensuring it succeeds over time.

Prediction #4 - More Kubernetes means more cloud and more open source projects

Any set of predictions for 2020 would be incomplete without discussing Kubernetes. Kubernetes is popular with developers, and that popularity is only going to grow. Kubernetes is still far in advance of the rest of the market around container management and orchestration. However, the challenge here is that the rest of the IT industry has to catch up and fill the gaps that do exist.

Behind this is the fact that Kubernetes does a great job of hiding complexity when it comes to managing containers. However, when anything goes wrong or needs tweaking, it's very hard to dive into. Consequently, more emphasis on the tooling and integration around Kubernetes will be needed.

This should actually be a boon for open source and for developers. The release of Kubernetes operators that provide the functionality to automate resource scaling should help integrate applications with projects. For developers, these operators make it easier to integrate with microservices-based applications, so these should be available under open source licenses.

At the same time, the scalability requirements for these applications will lead to more consumption of cloud. However, rather than betting on one cloud provider, Kubernetes will encourage more companies to look at hybrid and multi-cloud deployment options. Even those teams that don't plump for full cross-cloud deployments will see the benefit of a more competitive cloud marketplace.

##

About the Author

Patrick McFadin, Vice President of Developer Relations, DataStax

Patrick McFadin is the vice president of developer relations at DataStax, where he leads a team devoted to making users of DataStax products successful. He has also worked as chief evangelist for Apache Cassandra and consultant for DataStax, where he helped build some of the largest deployments in production. Previous to DataStax, he was chief architect at Hobsons and an Oracle DBA/developer for over 15 years.