At VMware, we believe that VMs and containers work best together to provide an agile, efficient infrastructure for application development and deployment. But no web-scale application runs on just a single VM, or even a small group of VMs. Application deployment and operation requires a well-managed cluster, with the ability to elastically scale to meet service demands. Just as Kit mentioned earlier at DockerCon, we are introducing a solution to better assist DevOps engineers with these challenges.
In September of last year, we introduced Big Data Extensions (BDE) for vSphere to enable enterprises with deploying, running, and managing clustered workloads like Hadoop. Because BDE was designed to help our customers easily deploy clusters, and plan and automate VM provisioning, we saw an excellent opportunity to extend vSphere BDE to provision and manage Mesos and Kubernetes clusters as well.
While we’ve previously written about our collaboration with Google’s Kubernetes project in this areas, today we are adding Mesosphere to our ecosystem using this BDE extension. The latest BDE fling enables vSphere users to stand up a Mesos or a Kubernetes cluster in minutes, via a simple GUI or a single spec file. The clusters, as one would expect, can be customized for size and topology, and to scale elastically on demand.
BDE with Mesos and Kubernetes integrations can be downloaded from our flings page, where we frequently release new technologies. We look forward to hearing about your experience in using this Mesos/Kubernetes integrated BDE package to deploy, manage, and scale your applications – come talk to us at @cloudnativeapps.
Jesse Hu is a developer for Project Serengeti (also known as Big Data Extensions) in the VMware Beijing R&D Center. He holds an M.S. in Computer Science and is interested in Open Source, Big Data, Cloud Computing, and Linux Containers.