The 1982 science-fiction film Blade Runner and a 2017 sequel Blade Runner 2049 explore the relationship between humans and the environment in a ‘retrofitted’ future that appears quite bleak. An ambitious new TM Forum Catalyst proof-of-concept project called Blade Runner is imagining a much brighter future where communications service providers (CSPs) can use 5G, augmented reality (AR) and multi-access edge computing (MEC) to deliver enterprise services end to end across partners’ networks.
AT&T, BT, du Telecom, NTT, Orange, Verizon, Telecom Italia Mobile and Vodafone Group are championing the project, which includes BearingPoint, CENX, EXFO, Infosys, RIFT.io and Riverbed as participants. Other companies also are considering joining the team. The idea is to show how CSPs can model 5G services and then expose them between operators using common information models and TM Forum Open Application Program Interfaces (APIs).
Vodafone is leading the project, drawing on lessons learned in a previous digital ecosystem Catalyst called Do Drones Dream of Electric Sheep? (a play on the Philip K. Dick novel Do Androids Dream of Electric Sheep, upon which the Blade Runner movies are loosely based). That project used 5G network slicing for internet of things (IoT) applications, including offering drones as a service for precision farming.
The goals of the Blade Runner project are more ambitious and include:
- Demonstrating a standardized way to share service catalogs and use open, well-defined APIs for common operations
- Putting the TM Forum Open Digital Architecture (ODA) to the test, showing how orchestrators at different layers of the network must work together to deliver services end to end
- Testing and enhancing TOSCA (Topology and Orchestration Specification for Cloud Applications), an orchestration data modelling language that OASIS is developing, by adding service assurance capabilities
- Testing and improving TM Forum and MEF APIs (for example, the TM Forum Product Catalog, Service Catalog, Service Order and Performance Management APIs, and MEF Interlude)
Using AR for remote repairs
At Digital Transformation World in Nice, France, in May, the Blade Runner team will demonstrate a complex and interesting enterprise use case using AR to help a local engineer in a Dubai mine repair earth-moving equipment (such as from Caterpillar or JCB) by receiving instructions from an expert located at an enterprise customer’s facility in Germany.
The equipment engineer in the remote and hard-to-reach mine is fitted with Microsoft HoloLens hands-free AR googles, which he uses to connect with the expert via a Skype call. The expert can see exactly what the engineer in the mine sees and provide clear, step-by-step instructions for the repair. Both expert and engineer can draw and highlight on top of a 3D rendering of the truck.
This video from Caterpillar shows the HoloLens goggles in action:
How does the Catalyst work?
For the purposes of the demonstration in Nice, the team will use Amazon Web Services (AWS) instantiations to simulate a Vodafone data center in Germany and MEC in a du base station in Dubai with 5G connectivity assumed. Attendees who visit the Catalyst on the show floor will get to try out the HoloLens goggles as part of the demo.
The goal is to demonstrate four end-to-end CSP processes:
- Service modelling and onboarding
- Dynamic service orchestration
- Real-time service monitoring
- Wholesale settlement
Below is a depiction of what attendees will see during the Catalyst demonstration, followed by detailed diagrams and a step-by-step explanation of the processes and how participants interact during each.
1) The engineer puts on the HoloLens goggles and turns them on.
2) The backend AR service captures geolocation information and sends this to a Vodafone customer relationship management (CRM) system.
3) Vodafone CRM then forwards the request to its service orchestrator (SO) to dynamically fulfill the end-to-end service chain.
4) Based on the location (Dubai), the Vodafone SO determines that du is the best partner to handle the request and that it’s necessary to use MEC in a local base station to give the equipment engineer in the field the best possible experience.
5) In the Vodafone data center in Germany, the SO configures the necessary virtual network functions – a firewall, software-defined wide area network (SD-WAN) and virtual probe (in this case, the SD-WAN capability is provided by Riverbed and the virtual probe is provided by EXFO).
6) The Vodafone SO communicates the service requirements to a du SO through an MEF Interlude API.
7) The du SO configures the MEC service, which includes not only firewall, SD-WAN and virtual probe VNFs but also the AR video capability (this will be provided by Infosys).
8) When the equipment operator puts on the HoloLens goggles, the service is activated and a Skype call is placed to an expert in Germany.
9) The services are constantly monitored end to end using EXFO probes and a service assurance system provided by CENX. As part of the demo, the team will introduce latency, which triggers a request to heal or scale out the AR function to ensure quality of service.
10) du issues an invoice to Vodafone for use of the AR service (BearingPoint provides this wholesale settlement capability).
“We are supporting closed loop assurance, and that is a challenge, having different service providers exchanging information about their service models,” says Dimitris Symeonidis, Enterprise Architect at Vodafone and leader of the Catalyst. “The concept is that we use TOSCA to define the service models in different layers.”
Putting ODA to the test
That’s where TM Forum’s ODA comes in. ODA is a blueprint for common operations and IT management, and one of its main principles is intent-based management, which abstracts the complexity of the network at a high level and then uses a customer’s intent, policy and machine learning to manage it. To make this work across partners’ networks, CSPs must agree to use the same information and data models along with APIs so that orchestrators in different domains can communicate.
The image below shows the Vodafone and du environments side by side and depicts the ODA layers, showing where each Catalyst participant fits and where Open APIs are required. It’s important to note that Vodafone does not dictate the specific technologies to be implemented by a partner provider. For instance, du might consider a different supplier of a firewall service than the one selected by Vodafone. Vodafone simply states the intent (i.e. the functionality expected), and unless otherwise specified in a service level agreement, the partner can pick any alternative in the same domain.
Orchestration at many layers
The Blade Runner team is aiming to demonstrate the need for the different orchestration layers proposed in ODA. Based on the architecture, the RIFT.io orchestrator will act as an end-to-end service orchestrator for the entire service chain. Each CSP will have its own instance of the product, abstracted through standard APIs.
The team also hopes to use the Open Network Automation Platform (ONAP) to act as a technology-specific domain orchestrator for the resource-facing services. This will provide the necessary abstraction for the technology environment, the virtual infrastructure manager (Openstack).
“We are considering different modules of ONAP including Service Design and Creation, a Master Service Orchestrator and a Controller,” Symeonidis says. “This will prove the potential capabilities of the product within the ODA architecture.”
Watch Vodafone’s Symeonidis and BearingPoint’s Dirk Rejahl discuss the Blade Runner Catalyst at TM Forum Action Week:
This is just the beginning for the Blade Runner project. In subsequent phases, the team would like to work on end-to-end service orchestration for large enterprise customers with multiple global locations and even more complex service-monitoring scenarios. Because the service involves multiple service providers working in tandem, it is a priority to get security and data governance right. They also want to explore business cases and benefits for real-time policy management and charging for cloud services, including self-healing and spot pricing, a concept that has been introduced by AWS but so far hasn’t been applied in a telco environment.
If you are interested in participating in the Blade Runner Catalyst, please contact Tania Fernandes via [email protected].