It's good news for the industry that the Linux Foundation and the European Telecommunications Standards Institute (ETSI) announced they had signed a memorandum of understanding (MOU) to more closely work with each other.
But why did it take so long?
ETSI and the Linux Foundation said the MOU would bring open source and standards bodies together while fostering synergies between them. Areas of collaboration between the two could include NFV, management and network orchestration (MANO), AI and edge computing, according to today's press release. All good.
ETSI and the Linux Foundation have worked with each over the years, but the fundamental difference, it would seem, was over ETSI's support of Open Source MANO and the Linux Foundation's backing of ONAP. While service providers such as AT&T and Orange were firmly in the ONAP camp, other large service providers, mainly Telefónica's, backed OSM. Telefónica's OpenMANO project became a building block for ETSI's Open Source MANO (OSM) code.
RELATED: Editor's Corner—Open source is not 'one size fits all'
OSM was created with the goal of providing a fully functional, open source NFV orchestrator to go with ETSI's NFV framework. ONAP, which was largely created out of AT&T's internally developed ECOMP project, is about more end-to-end automation and orchestration. Different scopes for different folks, but there has been some overlap and some friction.
"What we have here is actually two related activities that worked apart from the first, and worked under very different mandates," said Tom Nolle, president and founder of consulting firm CIMI Corp. "OSM has one focus, which is on the management of the VNF hosting process, and ONAP has the broader mission of service lifecycle management. You’d think we’d have wanted to start with the broad mission, and fit OSM into ONAP, but OSM came along first. How do we turn back time and do an alignment then?
"But there’s yet another level, which is whether either of the two are any good, and that’s where I think this collaboration could really go off the rails. We did NFV wrong. We did OSM wrong because we did NFV wrong, and we accepted some of the same wrong principles, mainly ignoring the TM Forum's work on event-to-process steering via a service data model, in both. So do we now cooperate to do the wrong thing consistently and call that 'collaboration?' I don’t think it’s even progress. What the collaboration really shows is that when you don’t get that high-level vision right, you can’t fix it by enforcing it uniformly."
Both Nolle and Roy Chua said the details of the colloboration between ETSI and the Linux Foundation need to be sorted out and made public. There needs to be concrete measures, according to Chua, founder and principal at AvidThink.
"I would say it trends positive, but I withhold judgment until we see more tangible elements of collaboration," said Chua. "Take for example, OSM and ONAP. We've heard from ETSI and LF that these don't compete, and that OSM's scope is narrower than ONAP, but with this MOU, perhaps the integration points will be formalized, and we can see how ETSI and LF expect them to interface with each other."
At the Open Networking Summit earlier this month, the Linux Foundation's Arpit Joshipura, general manager, networking edge, and IoT, said competition between open source groups and standards development organizations was a myth driven by the media. As an example of cooperation, Joshipura cited the large number of open source and standards bodies participants on a panel at OSN.
The push to hybrid networks using NFV and SDN initially led to fractured ecosystems as service providers and the legacy vendors pondered how to move to a software, cloud-based world while still serving millions of customers without any noticeable hitches in services.
The struggle for smaller vendors or startups with limited budgets has been centered on choosing the right open source group to put their engineering efforts and money behind. Knowing that work with one open source group will carryover or integrate with another is some piece of mind for those vendors.
The industry has seen a fair amount of fruitful collaboration over the past year. ETSI and Open Platform for NFV (OPNFV) collocated their Plugtest and Plugfest, respectively, testing platforms for the first time during an event last year in Sophia Antipolis, France. Both MEF and TM Forum contributed to ONAP's Beijing release. There are also hooks in ONAP that work with OSM.
Earlier this month, the O-RAN Alliance cemented its collaboration with the Linux Foundation, by creating the O-RAN Software Community (O-RAN SC.)
"It's encouraging to see how far the industry has come in such a short time," said Joshipura, in a prepared statement. "This agreement with ETSI signals it's possible to reach a harmonization of collaborative activities across open source and standards for the networking industry. Working together results in less fragmentation, faster deployments, and more streamlined innovation."
While any level of collaboration is better than no collaboration at all, Nolle made the point that it's hard to put the OSM and ONAP genies back in their bottles and start fresh.
"At the high level, it’s clear that you can’t have open-source software and standards that apply to the same technology targets happily running off on their own tangents, or at least one or possibly both activities will be a waste of time," he said. "Cooperation at this level is good, but the devil is in the details. In order for it to work, it’s logical to assume that standards would frame a high-level model that would be fleshed out and implemented via open source." — Mike
Editor's Corners are opinion columns written by a member of the Fierce editorial team. They are edited for balance and accuracy.