JFrog and Kubernetes. Our Cloud-Native Journey Continues

Here at JFrog, we are avid users of Kubernetes in our various development efforts and in our different environments. That’s why we’re proud and excited to join the Cloud Native Computing Foundation so we can now also contribute from our technology and know-how. Learn more about our membership in this recent press release.

JFrog benefits greatly from the robustness of Kubernetes and its operational model that is independent of underlying infrastructure. We operate our SaaS offerings on multiple public clouds, and our customers operate our products on many types of infrastructure. We are currently on a journey to move our SaaS offerings to Kubernetes in order to standardize operations across AWS, Azure, and GCP.

We provide a Helm chart for our on-prem offering for anyone who wants to run Artifactory in their Kubernetes cluster. You can download it from the public Helm GitHub repository.

On the contribution side, we believe our years of experience with the huge diversity of binary package types will add another valuable voice to the cloud native community. Through our upcoming Helm package support in Artifactory, we will provide the same traceability and tracking for Kubernetes Helm that we’ve provided for Docker images, npm, Maven, and many others. When you need to understand the impact of changes in the dependency tree, JFrog Xray enables the automated analysis of Docker images and their metadata to produce the long chain of dependencies that help you answer questions like, “what is in that WAR?”. The result? You will know exactly what you have running where, and where it came from. Xray can do this because it understands the metadata for the binary contents including licenses, vulnerabilities, and more. The emerging story here is that the software metadata is critical in order to have reliable release automation. So together with Google and others we’ve recently launched the open source Grafeasmetadata API (view the recorded webinar or read our blog post about it) to enable consistent sharing for all the information you need to understand your software supply chain.

Our Journey

Over the past several years JFrog has grown enormously in several ways. We have experienced growth in product adoption in the market, expansion of our product portfolio and also an increase in the pace at which we adapt to the dynamic nature of expediting software releases.  As with any rapid pace of growth, we have also experienced a few challenges to overcome along the way. Take our SaaS offerings for example. Initially Artifactory SaaS was only available on Amazon Web Services. Quickly after launch it became clear that we had to provide the same service on Microsoft Azure and Google Cloud Platform to address growing customer demand. These clouds differ in many ways, most significantly in APIs and operational models. The cost of maintaining SaaS offerings in three different clouds is also significant. All of these factors combined lead us to believe that we must become cloud native.

The question then arose as to what operational model would allow JFrog to manage cloud native services at scale across all the major clouds, provide the ability for customers to use the same operational knowledge, and support hybrid cloud service models that would ensure support for distributed work, High Availability, Disaster Recovery, etc. Kubernetes emerged as the most viable and promising option to overcome these challenges. Kubernetes provides the ability to scale globally, self-heal, and manage operational costs that we and our customers demand.

Cloud native is about speed. JFrog products are the secret weapon for ensuring safe, efficient, and expedited cloud deployments. Artifactory was created ten years ago, and therefore, wasn’t originally designed to be cloud native. Over the past year this has changed. Artifactory has evolved significantly, and is now packaged as a Helm chart for deployment on Kubernetes , regardless of where the Kubernetes cluster runs. Our other products, Bintray, Mission Control and Xray were built as cloud native microservices from scratch.

During this journey to become cloud-native, JFrog has recognized that the communities surrounding Kubernetes and the other Cloud Native Computing Foundation (CNCF) projects are a powerful force for innovation and digital transformation. We are excited to be a part of CNCF and support other open source projects through membership and through our initiative that offers Artifactory SaaS on GCP free for selected open source projects. We plan to contribute our learning from both our own transformational experience in the form of enhancements to Helm, and container lifecycles in Kubernetes, as well as to benefit from the tremendous insights of our partners in this community.

Keep your eyes on JFrog Artifactory Release Notes for the forthcoming release with full support for Helm charts.

Ready to start managing your the Docker images and all other packages that make up your application?