Lifecycle orchestrator could give operators the edge in enterprise 5G

A TM Forum Catalyst team, which is backed by seven operators, is developing a blueprint for an edge cloud lifecycle manager which could benefit the whole industry. The 5G enterprise market is set to be worth $19.5 billion by 2025, according to Zion Market Research. Many operators are targeting enterprise opportunities first to recoup return on investment (ROI) on their significant 5G investments to date. Opportunities exist in sectors such as retail, gaming and entertainment, transportation, government, manufacturing, healthcare, IT, agriculture and more. The ability to deploy and manage dynamic network slices in a fully automated ‘zero-touch’ way is a critical success factor for enterprise 5G. Edge cloud management is the other missing piece and is being tackled by a TM Forum proof-of-concept Catalyst project, titled Becoming EDGY. At the Catalyst awards in Digital Transformation Asia last week, the project won the prize for best new Catakyst in show. Find out more in this article. Backed by the industry The project has seven telco champions: AIS, Airtel, Globe Telecom, KDDI Research, Singtel, Optus and Telkomsel, highlighting the work’s importance to the sector. Catalyst champions bring a business challenge that they need to solve, and each offers a unique angle based on their geography, strategy and more. Champions work alongside participants – in this case, Ericsson, Infosys, Intel, HPE and VMWare – who each provide a technical piece of the solution. Communications service providers (CSPs) need to be able to ensure that the 5G use cases for enterprise vertical market markets can be implemented quickly, at speed and scale, and that they can be managed consistently. “We need to be able to offer new services in a matter of weeks to service the enterprise market,” said Geoffrey Ganal, Enterprise Architecture Group Project Lead, Globe Telecom. “To do this, we have to leverage edge computing to deploy the use cases and ensure that the user experiences is optimal.” Edge computing allows data to be captured, processed and analyzed close to where it is created to enable real-time analysis and actions and to reduce latency. Ganal explained: “Edge use cases have varying characteristics and thus require various edge implementations in order to be efficient and cost-effective. Being able to scale beyond public, private and hybrid clouds extending automated lifecycle management to the edge is important.” He added: “There is currently no standardization in this space and many open source projects have been created to address the requirements. This Catalyst seeks to explore the maturity of the solutions in the market for edge cloud management and how this should be wired up consistently with the automation and orchestration happening close to the core sites.” The Catalyst team recommends that the solution stack used at the central data centre for infrastructure lifecycle management should be extended to the edge, thereby reducing the need for new solutions in this space while minimizing the cost of operations. “Some edge solutions e.g. mobile edge computing (MEC) and fog computing would require pushing the virtual infrastructure management to the edge, while some would require the central data centre’s infrastructure management to directly manage the application workloads at the edge e.g. [deploying a] cloudlet,” said Ganal.

On show

At Digital Transformation Asia (12-14 November, Kuala Lumpur), the team demonstrated how their proposed blueprint for an edge cloud lifecycle manager could work. They showcased the solution’s value in several sectors, including a video analytics application in a retail store to help companies make more informed choices about store layout and product placement. Another example showed how a gaming company could deploy compute power for the gaming server close to the customer premises, instead of in a central cloud, ensuring gamers will experience almost real-time processing and no jitter – particularly important for graphics. Applications for content delivery network (CDN) hosting will also be showcased. “The focus is on how the workloads of each of these edge locations can be managed centrally from the lifecycle service orchestrator,” Ganal said. “Further, we will demonstrate that this lifecycle manager is applicable to different edge implementation approaches.” Each company in the team plays a key role. Ericsson contributes a virtual infrastructure manager; Intel provides its OpenNESS Edge Accelerator; VMware provides its Integrated Openstack and NSX-T; HPE provides edge hardware; and Infosys contributes end-to-end system integration for edge hardware. The team has used a number of TM Forum assets in the solution, include the following Open APIs: Service Catalog, Shopping Cart and Service Order. They will contribute feedback and new extensions back to the community for wider benefit.