It would adapt and leverage a huge, core asset by turning it into an enabler. Right now, many telcos have begun digital transformation, but projects often happen in silos, including those to change OSS/BSS. How do we get from here to there, is the question TM Forum’s members are striving to answer. The siloed approach results in incremental improvements whereas radical change is needed if operators are to compete against (and partner with) digital natives to deliver services customers really want.
“We have to look at how quickly we can respond to opportunities, because we’re not just competing with other service providers, but also with internet giants,” TM Forum CEO Nik Willetts told members gathered here in Vancouver at Action Week. “Perhaps even more frightening, we’re competing with companies that don’t even exist yet. We know something has to change, and it’s more than just incremental improvement.”
What’s your vision for the future?
He added, “We have to ask ourselves the question that was posed to me at our smart city event in Yinchuan last week: ‘What would a company like Amazon do if it were facing this challenge? “I think we need to be dramatically more radical. We need to change the mindset of how we view the role of technology in the business itself.”
That means not looking at a telco’s enterprise IT assets (OSS/BSS) simply as a cost center, but as a strategic part of the business. “When you ask Google or Amazon ‘What does IT do for you?’, they say IT are just the guys who come and put a laptop on your desk, because for them IT is the heart of the business,” Willetts explained. “We need to be asking what IT can do for the rest of the business… Which parts of the business can we automate? Which parts of our data pools can we use to inspire new products? What capabilities are we sitting on that we could be monetizing but are not?”
The future of OSS/BSS
In light of these challenges, TM Forum has introduced a new architectural vision called the Open Digital Enablement System (ODES) which reimagines OSS/BSS. The graphic below shows the key drivers for ODES, which include network virtualization and the move to cloud; the need to enable digital ecosystems connecting partners, particularly for IoT; and a requirement to automate operations because the volume and velocity of changes that will occur in a software-defined network simply can’t be handled with manual processes.
What is ODES?
The ODES architecture (see graphic below) was developed with input from many of the Forum’s key CSP members during a series of workshops, and is consistent with Frameworx definitions and domains. It uses existing TM Forum assets including Open APIs and key concepts from the Zero-touch Orchestration, Operations and Management (ZOOM) project’s work on the Hybrid Infrastructure Platform. It also draws heavily from work on OSS/BSS of the future. You can find a more detailed introduction to ODES in in this blog written by Brian Levy, Strategic Advisor to the Forum. TM Forum members also may access a white paper about ODES here.
What do operators need?
We solicited feedback from members about the proposed overarching OSS/BSS architecture during a day-long workshop at Action Week on Tuesday. More than 60 TM Forum members, including executives from AT&T, BT, Globe Telecom (SingTel), Microsoft, Orange, Telefónica and Vodafone, as well as many consulting companies and suppliers, attended.
CSPs need to transform monolithic legacy systems in an agile way, with loose coupling and cohesion being key requirements, according to Laurent Leboucher, Vice President, Architecture, Enablers and Security, Orange.
“A lot of operators’ IT transformation projects either have failed or almost failed – by almost failed I mean that if it takes more than two years to transform a BSS stack, it’s failed,” he said. “When you transform IT, it’s not just about defining the target; it’s really about how you can execute it in a reasonable timeframe that can achieve results in a very iterative and agile way.”
George Glass, Chief Systems Architect, BT, agreed, saying that CSPs’ overall OSS/BSS architecture is “much too big and complex,” and that operators need common components with granularity to enable flexibility in developing and delivering services.
BT has done a significant amount of work transforming it’s OSS/BSS already. Glass gave an example of billing system transformation that required exposing internal billing services to customer relationship management systems. Through this transformation, BT has been able to save about $4 billion over the past seven to eight years.
Overall, operators who participated in the workshops to develop ODES said OSS and BSS must be:
- standards-based, enabling a marketplace of commercial and open source innovation;
- unified in a single architecture;
- data-centric with a single data lake to enable automation and AI;
- platform-based and componentized for agility using TM Forum Open APIs;
- artificial intelligence (AI) capable and autonomous;
- real-time in order to respond to the expectations of customers; and
- ecosystem capable to allow easy interaction with partners.
During the Action Week workshop, CSPs, suppliers and systems integrators engaged in a spirited discussion of each of these areas of focus. Overall there was significant agreement on the vision and requirements, with many suggestions and ideas for refinement.
Following are some of the most important areas of focus and questions CSPs and suppliers would like to see ODES explore in more depth:
- Defining OSS/BSS and deciding how to combine them – several workshop participants pointed out that there really is no agreed definition within the telecommunications industry of what OSS is and what BSS is. What’s clear is that in the future they need to be one combined system, even if their functions are grouped as OSS and BSS during the transition.
- How will Frameworx be used within ODES? Frameworx has been used as the language of OSS and BSS design for many years. Going forward, we will need to be able to componentize Frameworx and bundle elements from each individual framework into a package. In the ODES white paper we have proposed calling these components ‘Framelets’.
- What’s the role for AI? – many participants urged caution against becoming overly excited about the potential for AI and machine learning. The industry needs to “walk before it runs,” they said. In addition, CSPs must determine how they are going to store data, whether to use a single data lake, for example, and decide whether this is something that needs to be standardized for all operators.
- How do we evolve the common language? As an example, the word ‘service’ is used at multiple layers in ODES. Frameworx already provides formal definitions, but we need to understand whether they need to be modified in light of the ODES layering. In addition, lots of new terms like ‘microservice’ need a clear definition. Several members contributed their definitions for consideration.
- How does the work of other standards-development organizations and open source groups affect ODES? This question came up several times during the workshop discussion. Compatibility with open source groups, in particular, was a key consideration for the initial work on ODES, and some of the leading CSPs involved in Open Network Automation Platform (ONAP) and Open Source MANO (OSM) contributed to the alignment. For more about the relationship between TM Forum and ONAP, see this article about the Linux Foundation’s Arpit Joshipura’s keynote at Action Week earlier this week. If you’d like to learn more about ODES, please contact Barry Graham via [email protected].