Kubegrid Service Automates Kubernetes Deployments
A Kubegrid service that enables IT teams to automate the deployment of Kubernetes clusters and the applications they run on onto public clouds is now generally available.
Kubegrid CEO Spartak Buniatyan says the company’s namesake service abstracts away the complexity of deploying containerized applications on Kubernetes cluster by first leveraging application programming interface (APIs) to spin up Kubernetes clusters on a cloud infrastructure and then packaging the application as a Docker container to be deployed using a GitOps model to automatically deploy any changes made to a Git repository to the Kubernetes cluster.
Based on cloud-native tools that include no proprietary code, the Kubegrid service makes it easier for mere mortal IT professionals to deploy containerized applications without have to master all the nuances of instructions provided by the team of software engineers building the core upstream version of Kubernetes overseen by the Cloud Native Computing Foundation (CNCF), Buniatyan says.
Kubegrid, he notes, will appeal to developers and IT operations teams alike because it enables specifications regarding clusters and cloud service providers to be declaratively defined from within a Git repository. That capability can then be used to create workflows to facilitate a continuous delivery strategy spanning pre- and post-production deployments, or just stay current with the latest version of Kubernetes without having to wait for an IT vendor to update a curated distribution of Kubernetes, he says.
In addition, those teams can use Kubegrid implement a automatic failover workflow spanning multiple instances of Kubernetes running on different clouds, Buniatyan adds.
Finally, to address potential cybersecurity issues, Kubegrid provides monitoring and continuous scanning of containerized applications via an alliance with Cloudflare.
Buniatyan says it’s more than apparent that the complexity of deploying and updating cloud-native applications on Kubernetes clusters is slowing down adoption in enterprise IT environments. Most enterprise IT organizations have stood up one or two Kubernetes clusters in a production environment. But the management challenges associated with managing Kubernetes deployments at scale is clearly limiting the number of Kubernetes clusters that enterprise IT organizations want to support in a production environment.
It’s not clear to what degree IT organizations will embrace and extend GitOps processes. But as developers continue to rely on Git repositories to manage application development, it only makes sense for IT operations teams that have embraced DevOps processes to share access to the same repositories.
Buniatyan also notes that many of those same enterprise IT organizations are also concerned about becoming inadvertently being locked into one cloud platform versus another by becoming overly dependent on proprietary tools. By relying on Kubegrid they can be assured of being able to seamlessly move workloads between multiple public clouds.
There’s no doubt that Kubernetes clusters are more than a little intimidating for the average IT administrator. As Kubernetes continues to evolve, however, there should be more frameworks and services becoming available to take advantage of Kubernetes application programming interfaces to abstract away many of the complexities of a platform that was clearly built by software engineers for other software engineers.