OSM Zooms Past First Development Milestone With Release Zero Code

Back in April, the coalition behind the development of the Open Source MANO (OSM) code made this rather bold public proclamation:

The ETSI OSM group announced leadership structure and roadmap for first specification release in 2 months, subsequent updates every 6 months

Meeting this goal certainly seemed easier said than done considering that the formation of the group was only announced in late February and there were no less than 23 companies involved at the inception of the project. Today the list of companies involved has increased to 35 with that number growing as more leading vendors and operators lend their support.

Open Source MANO LogoAs a founding member of OSM, we at RIFT.io are certainly pleased with the burgeoning ranks of companies. But we are prouder of the fact that the OSM has met the first stage of the admittedly aggressive roadmap with the introduction of the “Release 0” code as announced last week. This is an important, foundational step toward the introduction of the “Release 1” code, which will be a full-blown open-source community initiative.

Release 0 was made possible by integrating seed code contributions from RIFT.io along with Telefonica, Canonical and others, which helped OSM hit the ground running. The ETSI OSM team demonstrated the possibilities of this code at Mobile World Congress this year. Since then, the development team has been heads down on some key, operator-defined improvements to the features and functionality of the code.

One such advancement has been in improving the user experience and the ease by which virtual network functions (VNF) providers can develop data models for on-boarding in OSM. Both improvements are close-to-the-heart for us at RIFT.io, as they directly support our company’s stated mission. We are, of course, taking an active part in the effort. Indeed, our seed-code contributions to the OSM focused around network service orchestration (including easing VNF on-boarding) as well as the graphical user interface that provides a single pane of glass for configuring and reviewing the state of the orchestration environment and cloud infrastructure.

For the OSM code specifically, Release 0 presents a simplified on-boarding process by developing a single entry point for VNF and network service (NS) packages. The VNF and NS descriptors are now easier to read with the move to YAML format. Other improvements include moving embedded scripts in the descriptors into packages and separating out default values and initial configuration.

There were other, practical considerations that the OSM community put significant effort behind to make the code more user friendly for developers and operators. This includes the work to build a comprehensive set of publicly available documentation covering the data model in detail, the minimal infrastructure requirements, installation guides, how-to guides for users and developers, technical briefings, videos, the OSM tool suite, and the Software Development Lifecycle. It’s important to note that OSM achieved all this while ensuring that the end-to-end service instantiation demo shown at MWC’16 could be reproduced with the Release 0 code.

OSM is now using its Release 0 success as the platform for Release 1 code, which is expected to be available in six months. Specific projects in Release 1 will include further developing the OSM data model as well as ongoing enhancement to service and resource orchestration and VNF configuration and abstraction.

Meeting this next roadmap milestone will further help distinguish OSM from competing projects while getting the industry one step closer to an open source MANO stack using best-in-class open source workflows and tools to ensure rapid development and delivery. Time is certainly of the essence with operators becoming increasingly eager for fully workable NFV and NFV MANO solutions.

Based on how OSM met its first development goal, we have the highest confidence that Release 1 will be on-time and as advertised. See you in six months.