This is the second part of a two part blog about SEBA/VOLTHA. The first blog centers on background and accomplishments to date related to ONF’s broadband access initiatives; this blog outlines exciting development plans going forward.
Looking ahead, as the early adopter operators scale their broadband deployments multiple orders of magnitude to millions of subscribers and the vendor ecosystem engages with new operators, as is natural, there are new demands being placed on the SEBA/VOLTHA project for added functionality, performance and scale. A significant increase in automation of operations, deployment and troubleshooting is targeted to further reduce operational costs. In addition, there are active requests from the membership for ONF’s open source software stack to be more interoperable with standards/APIs defined by other industry bodies like the Broadband Forum. As we scoped the direction of SEBA/VOLTHA going forward, these considerations were taken into account.
To support existing deployments, expand their scale and increase the operator’s footprint, the SEBA/VOLTHA project is exploring focus on several areas going forward:
- Thread 1: Expand the scale, features, automation and testing,
- Thread 2: Aligning with Broadband Forum on outwards facing APIs
- Thread 3: Creating an open source, software defined and disaggregated BNG solution.
For SEBA/VOLTHA to deliver on its vision and enable the level of broadband access transformation that it has initiated, an essential level of joint development is envisioned to accelerate for global adoption. A number of targeted improvements and enhancements have been identified by the VOLTHA team to this end. This work covers a selected balance between scale improvements, bug-fixes and new features. These opportunities can be categorized into three threads:
Thread 1: Scale, Expanded Feature Set, Automation, CI/CD
In 2022 DT and TT deployments are scheduled to scale to hundreds of thousands customers. During 2022, additional major operators have plans to deploy VOLTHA extensively through the help of our software integrator partners. First and foremost we want to scale VOLTHA 10x from the existing 10,240 ONUs in POD to 102,400 ONUs.
To achieve such scale results, and to continue on the cloud native path and further reduce VOLTHA’s deployment footprint in terms of RAM,CPU and storage, ONOS classic should be removed in favor of a purpose-built golang based controller, based on µONOS, directly calling VOLTHA’s core API, removing the overhead of the OpenFlow protocol and the of-agent as well.
A byproduct will be to allow for easier deployments on an expanded plethora of options, including installing VOLTHA directly on the OLT, in the operator’s edge cloud or in the public cloud.
Expanded Feature Set
Operators have itemized various enhancements needed over the next year. Further expanding the feature set will allow deployment in more scenarios and with more operators. Features for joint development include:
- DPU Support,
- LACP support
- Additional operator workflows (e.g. TIM)
- Combo-PON OLTs with runtime configuration
Automation, CI/CD, DevOps & Test Suite
Extensive testing and more LTS releases underpin operational deployments, giving even more assurance of the stability and performance of the code.
Alarm notifications and automated performance measurement, collection and display through Grafana and Prometheus will allow operators to immediately visualize issues and diagnose the network. Feedback from operational deployments will be included into the feature list, especially including large scale deployment automation tasks.
All the new features also require work in the test suite, with dedicated tests to cover the additional code paths. Tests for existing functionality will be expanded and enhanced.
Thread 2: BBF Cloud-CO Alignment
Collaboration with the Broadband Forum (BBF) enables VOLTHA and ONOS to be integrated into the Cloud-CO architecture, alongside BBF’s solution and proprietary BBF compliant OLTs, thus allowing operators a wide plethora of choices for disaggregated OLT solutions, helping to overcome vendor lock-in.
ONF’s VOLTHA team plans to tighten the relationship with BBF, starting with collaboration around several working text and yang models (WT-454, WT-477, TR-383 and others), all revolving around disaggregated OLT functionality. After agreeing on the functionality and APIs, the goal is to make VOLTHA compliant with these APIs through a translation layer from BBF APIs to existing VOLTHA/Controller APIs, thus making VOLTHA a reference implementation of BBF APIs.
Thread 3: SD-BNG
A second collaboration point with BBF will be Software-Defined BNG, where the TASSEN group from ONF should come together with BBF to first discuss and then implement a cohesive API with a possible reference implementation in open source of a SD-BNG solution.
Furthermore the SD-BNG works aims at providing the first fully open source disaggregated and standardized BNG solution, offered on commodity hardware.
Under consideration is the idea to prototype an implementation on P4 hardware supporting the APIs to be defined. This would allow the APIs to be vetted through implementation (rather than just defining standards in isolation), and would provide a path for creating a powerful P4-powered BNG and perhaps even a collapsed broadband+mobile user plane.
ONF plans to advance SEBA/VOLTHA and further innovative capabilities into 2022 and beyond by leveraging the ONF’s unique model of blending engineering resources from ONF along with engineers and other resources from aligned companies in the ONF community.
We invite you to get involved in this exciting endeavor; please contact Andrea Campanella (firstname.lastname@example.org) if you would like to learn more.