Fresh off raising an additional $8 million in funding, Rafay Systems has made generally available a namesake software-as-a-service (SaaS)-based set of tools for managing Kubernetes environments at scale.
Rafay CEO Haseeb Budhani says it’s become apparent IT organizations need a SaaS-based approach through which they can centralize not only the management of Kubernetes clusters but also all the tools required to manage the Kubernetes environment.
As part of that effort, Rafay has created an application abstraction framework to define and manage the life cycle for containerized applications end to end. In effect, Rafay provides a system of record for all the instances of Kubernetes deployed, he says.
The Rafay platform also makes it easier to manage Kubernetes multiple clusters at scale from the cloud to the edge, capturing configuration blueprints that can be reused, Budhani notes. It also enables IT teams to deploy complementary tools to address requirements such as secrets management, runtime configuration updates and log and metrics.
Rafay extends the application programming interfaces (APIs) for both native Kubernetes distributions and managed Kubernetes services to create a higher level of abstraction in place of scripts and YAML files. This now makes it possible to manage multiple Kubernetes environments from a single SaaS-based console, he notes. In many cases, he adds, IT operations teams are too intimidated to master the many different scripts and YAML files, and this is holding back organizations from deploying Kubernetes in production environments.
Prior to launching Rafay, Budhani and Hemanth Kavurulu, vice president of engineering for Rafay, founded Soha Systems, which Akamai Technologies acquired in 2016. Akamai today sells the Soha technology as its Enterprise Application Access (EAA) platform.
While many organizations have trained teams to manage IT infrastructure as code, there are many more that rely on IT administrators who still prefer to employ graphical user interfaces to manage IT environments, as they don’t have the programming skills required to master APIs and YAML files. Rafay is betting that there will be a lot of DevOps teams that might simply prefer to use a SaaS-based console to manage Kubernetes at scale versus having to spend time setting up individual clusters.
The one thing everyone can seem to agree on is that managing IT operations has become more complicated. Without increasing reliance on IT automation frameworks, it won’t be possible to manage Kubernetes environments at scale. In fact, without that ability, it’s hard to see how the number of containerized applications deployed in production environments would increase. Almost every organization has a handful of Kubernetes clusters deployed, but the overall percentage of enterprise applications deployed on Kubernetes clusters in production environments remains slight.
Developers will, of course, continue to push to develop containerized applications that they perceive to be both more agile, robust and secure. The challenge IT organizations now face is finding a way to get IT operations teams to share in that enthusiasm.