Docker Advances Distributed Application Portability

By CIOReview | Wednesday, July 8, 2015

SAN FRANCISCO, CA: Docker, a company behind the Docker open source platform releases a new multi –host Software Defined Networking (SDN), a Docker Plugin Architecture and also enhances its three orchestration tools.

Docker multi-host SDN will be accessible as native platform functionality to assure seamless communication of multi-container distributed applications across IP networks, along with portability across any network infrastructure. Binding Docker’s new SDN functionality directly with pre-existing standards the DNS (domain name system) and VXLAN (virtual extensible LAN) ensures interoperability for multi-container applications and also between Dockerized apps and legacy applications. The DNS makes sure that Dockerized services will be able to communicate without modification. VXLAN enables the creation of portable, distributed networks that allow an application’s micro services to reside on any member of a Swarm, a native Docker cluster.

“Individual developers, through a single command, can establish the topology of the network to connect discrete Dockerized services into a distributed application. And then through a set of commands be able to inspect, audit and change topology ‘on the fly’,” says Solomon Hykes, CTO and Chief Architect, Docker.

Docker has created a new plugin architecture by collaborating with its ecosystem partners- CLusterHQ, Glider Labs and Weaveworks. The plugin offers an SDK (Software Development Kit) model for discrete areas of integration in which a native Docker SDN can be replaced with a third-party product.

New individual feature enhancements have been added to Docker’s three orchestration tools Docker Machine, Docker Compose and Docker Swarm. The area of functionality enhancement is in accord to the new multi-host networking capability. Through integration with Docker Swarm, the multi-container application can be immediately consolidated across multiple hosts and can communicate seamlessly across a cluster of machines with a single command.