Transport Network Management: A Multi-Vendor / Multi-Layer Case Study


Transport Networks: The Situation

Lumina recently started working on a transformation project for a Tier 2 telco in Europe (we’ll call them Telco X) with a pretty typical challenge - incorporating their multi-vendor WAN into the end-to-end network automation designs.

Transport networks typically include an array of vendor-proprietary tools. Within this closed environment, optical network equipment is managed by vendor-specific proprietary Element Management Systems (EMS). With every vendor, device, proprietary model, and management protocol the challenges streamlining management of this varying environment increases. 

Transport network devices from one vendor to the next have different management protocols for the same use case. This means that while one device will use EMS-driven proprietary L0/L1/L2, another device is netconf-yang compatible with proprietary yang. As networks increase in size to fulfill the increasing demands, this complexity continues to increase as well as the Capex and Opex. In this case, Telco X needed to provision services and compute bandwidth across a transport network which includes several models of Ciena, Infinera, and Adva resources.

Integrating packet and optical networks, specifically, creates unique hurdles for a transforming network. Integrating these two types of networks requires specialists since no standardized approach exists for managing this equipment. The separated control of the two domains makes change management a challenge—activities are complex, potentially taking several days or even weeks to implement. Also because of separate domain control, fault management is a challenge with limited communication between the layers. All in all, by keeping domain controls separated, service quality and customer experience is negatively impacted.

The Goal: Standardizing Control of Un-Standardized Network Elements

With this project, Telco X is working toward creating open, standards-driven optical network interfaces. New standards they wanted to drive to these previously closed network elements include OpenROADM, ONF T-API, and OpenConfig.

Parallel to the optical interface standardization, Telco X wanted to disaggregate optical devices (e.g. use whiteboxes). Good examples of the progress being made to disaggregate are the Telecom Infrastructure Project (TIP) OOPT as well as the TIP and Open Network Foundation (ONF) collaborative project, ODTN. These initiatives highlight the importance of Xponder disaggregation—critical as CAPEX is higher for Xponders than ROADMs. The initiatives also address Device disaggregation which requires a standard interface (e.g. OpenConfig) to configure the device.

Before implementation of the below solution, Telco X managed these varying devices with the corresponding Network/Element Management Systems (NMS/EMS) or directly through OSS/BSS. With addition of each vendor in the network, network management had become a complex function—too cumbersome to manage manually in a digital era.

The Solution: Service Automation

Telco X selected Lumina Networks’ Service Automation solution which is industry-proven, with large-scale production deployments.

Solution Features:

  • Pure play open source solution based on OpenDaylight solution features
  • Ubiquitous control layer hardened to enable carrier-grade performance and scalability 
  • Unifies management plate for heterogeneous network devices—simplifies workflow automation and resource control with a wide array of standardized southbound & northbound interfaces
  • Extensible using a microservices architecture 
  • Acting as a universal translator, shields the complexity of underlying southbound interfaces and enables northbound applications—allows NetOps teams to develop in any language they prefer
  • Agile development and deployment strategies—CI/CD and key updates

Solution Components:

Lumina’s Service Automation solution is built on the foundation of Lumina’s Extension & Adaption Platform (LEAP), which extends the benefits of Lumina’s OpenDaylight SDN Controller. The solution is highly scalable and microservices-based. The solution includes pre-configured defaults to streamline integration, but is easily customizable.

Common models (YANG):

  • Reduce OSS/BSS integration costs common models-yang
  • Single Northbound REST Interface
  • Leverage standards (Openconfig, TAPI, OpenROADM, IETF, …)
  • Rapid service deployment


Service Mapper: enables model to model translation

  • Translates common to specific models (NBI - SBI) Service Mapper - enables model to model translation
  • Configuration driven (JSON)
  • Apply defaults, morph values, …




Workflow Engine: Enables customized use case 

  • Pre/Post checks plus error handling Workflow Engine - Enables customized use case
  • Abstracts complex provisioning
  • User defined with external script support




The Outcome & Benefits:

  • Improved Customer Experience
    • Rapid new network & service feature delivery (days/weeks vs. months/years)
  • Decreased Costs
    • Improved network quality & utilization
    • Standardized management of multi-layer, multi-vendor, multi-domain packet/optical network equipment
  • Increased Revenue Opportunity
    • Makes transport resources dynamic and visible to applications for newer services
    • Enables NetOps and DevOps innovation
  • Simplified Operations
    • Make non-standard devices/VNFs appear as standards-compliant
    • Centralized change history and easier audit / PCI compliance
    • Automated multi-layer restoration

To see how a demo of how LEAP can enable your network, please contact us.

Programmable Data Center Fabric for Hyperscale Networks

Additional Resources:

Intent-Based Service Automation


Lumina SDN Controller