Telcos know they have to increase efficiency significantly to remain competitive, but the question is how? There is plenty to streamline: their complex infrastructures, legacy ways of working, and operational and management silos.
Transformation in these areas is almost like a mantra for those championing change, but making change happen is another thing. A new TM Forum Catalyst project called Unified digital infrastructure, which is being demonstrated at TM Forum Live! Asia in Singapore this week, seeks to “ensure optimal deployment and utilization of infrastructure resources while taking full advantage of cloud-computing properties,” according to Vincent Seet, Head of Enterprise Architecture at Globe Telecom, a champion of the project.
What’s the problem?
“This Catalyst is relevant for any aspiring DSPs [digital service providers] who have traditionally managed their infrastructure cloud resources by how they were separated – for example, three separate departments, one being enterprise business, another IT and the third, network groups,” Vincent says.
When these three separate departments acquire resources but don’t coordinate with one another, the resources they purchase can overlap. They may unnecessarily buy and manage resources that already exist in their company and are already being managed by the other departments. This means there’s a higher total cost of ownership (TCO) with more resources, manpower and skillsets used to operate the disparate infrastructure.
Catalysts are R&D projects where TM Forum members – both large and small companies – collaborate using TM Forum best practices and standards to develop innovative solutions to common industry challenges. The Unified digital infrastructure Catalyst in particular seeks to demonstrate the feasibility of doing away with silos and running network, IT and enterprise workloads on common digital infrastructure, and managing these workloads uniformly across multiple data centers.
Here’s an explanation of the different workloads:
- Enterprise workloads – typically software as a service (SaaS), virtualized or cloud applications such as virtual customer premises equipment (vCPE)
- IT workloads – referring to cloud applications such as databases, middleware, business support systems (BSS) and operations support systems (OSS)
- Network workloads – including virtualized network functions such as virtual evolved packet core (vEPC) and virtual IP media subsystem (vIMS)
“The vision of establishing a unified digital infrastructure agnostic of hosting all kinds of workloads requires a high level of maturity and openness in the NFV [network functions virtualization] cloud ecosystem,” Vincent says. “This allows these components to interoperate seamlessly, thereby allowing service providers to have true freedom of choice on any combination of such solutions.”
Globe Telecom and Telefónica are championing the project, with both sharing the same vision of a unified digital infrastructure. As CSPs, they provide the requirements for the project. Participants include:
- Amdocs and Ericsson providing virtualized workloads, NFV/cloud orchestrator, lifecycle service orchestrator, BSS and marketplaces
- DellEMC and Ericsson providing underlying hardware infrastructure
- HCL Technologies providing independent end-to-end testing and validation of the solution
- VMware providing the virtual infrastructure manager, the hypervisor and virtualized workloads
The integration architecture of these provisions works as follows:
Making use of TM Forum assets
The team implemented processes outlined in TM Forum’s Business process framework (eTOM) to ensure end-to-end zero-touch orchestration of orders or subscription workloads by end customers. They also used guiding principles from the Zero-touch Orchestration, Operations and Management (ZOOM) project and its Hybrid Network Architecture.
What can TM Forum Live! Asia attendees expect to see?
At TM Forum Live! Asia participants will explain the common, organization-wide fabric they have developed, and show how this has helped to pool infrastructure resources and manage them uniformly.
The hardware infrastructure provided by Dell EMC and Ericsson are co-located in the Dell EMC solution center in Singapore. Hypervisors, VIMs, orchestrators and virtualized workloads software provided by Amdocs, Ericsson and VMware are installed on the hardware infrastructure. Both service catalogs of the NFV/cloud orchestrators of Amdocs and Ericsson contain the same list of workloads that can be integrated. The audience will be able to choose a combination of any of the orchestrators, VIMs and hypervisor solutions to integrate their desired workloads.
The following illustration shows how deployment will be demonstrated at the event:
Source: TM Forum
“In the end, [DSPs] need to ensure optimal deployment and utilization of infrastructure resources which take full advantage of cloud-computing properties such as hyperscaling,” according to the Catalyst team. “In turn, they can launch new offerings to the market quickly, attain service agility and stay ahead of the competition. [DSPs] will see success when overall TCO and cost to serve are reduced tremendously.”