This content is part of the Essential Guide: The complete rundown on Docker data storage and containers

Hyper-convergence for Docker data containers debuts

Startup Rancher Labs builds out its offerings for Docker data containers with new persistent storage services and a hyper-converged platform for application developers.

Rancher Labs recently introduced a hyper-converged infrastructure platform for Docker data containers combining its operating system and the startup's new Persistent Storage Services for Docker. The platform gives developers the ability to deploy storage with containerized applications.

The company, which last summer announced it raised $10 million in funding, offers a lightweight Linux-based operating system called RancherOS used to run Docker containers. Along with Persistent Storage Services, the stack also includes the Rancher management platform to run Docker data containers.

Rancher Labs demonstrated the hyper-converged system at the DockerCon Europe conference this month and is in beta testing for the product.

Rancher runs virtual machines inside Docker data containers, allowing customers to use the same tools to manage both. Rancher's Persistent Storage Services are built on Docker 1.9 volume plug-in capabilities so that storage services run directly on container hosts. The service can create and mount persistent Docker volumes for applications and offer vendor-specific storage capabilities such as snapshots, backup, remote replication and data analytics. The Rancher platform runs NexentaEdge, Gluster and Ceph on its hyper-converged platform, and CEO/founder Sheng Liang said more storage will be added.

"Our persistent services are defined to use Docker Compose," Liang said. "You can deploy Gluster, Ceph and NexentaEdge to form reliable storage clusters. If the application requires remote replication, then they look for that piece of the technology that supports remote replication, maybe Gluster, and deploy it on the application."

The hyper-converged infrastructure for containers uses Intel x86-based servers and solid-state drives. Developers can use native Docker command line interface and API, and expand capacity by registering instances in public clouds as computing resources. It supports Docker tools Compose and Swarm, and Google Kubernetes.

Learn how virtual machines and Docker containers can coincide, and what this means for users.

The persistent storage services can be distributed across disks running on different hosts, either virtual machines or bare-metal servers, within the Rancher cluster. Rancher Labs can orchestrate the deployment of a block storage volume from the Gluster, Ceph or NexentaEdge service.

The persistent storage services use Docker Compose templates and they are launched from the Rancher UI or API. Rancher service manages both the application and storage services, monitoring any failures, and automates the recovery of storage services if there is a problem.

Stephen Hendrick, principal analyst for application development and deployment research at Enterprise Strategy Group, said Rancher container management is like "Docker in a box." He said the main advantage of containers is they give micro-service support in a way that traditional infrastructure data centers cannot.

"Even with virtual infrastructures, you still can't get the level of utilization that you want," Hendrick said. "Using persistent storage services is a way to provide direct native connection to different storage services to users. It eliminates the API mapping level; therefore, allows a more direct storage service that the enterprise is using."

Next Steps

Coho Data upgrades DataStream software for Docker containers

EMC adds plug-ins for shared Docker storage

Best practices for Docker container backup

Dig Deeper on Hyper-Converged Vendors and Products