East-West Interfaces – Horizontal OSS Coordination
Learning Objective: Understand East-West interfaces – horizontal coordination between peer or federated OSS domains, orchestrators, and domain controllers. Critical for multi-domain orchestration, 5G slicing, and O-RAN architectures.
The Three OSS Communication Directions
East-West interfaces enable horizontal coordination between peer or federated OSS domains and orchestration components.
What Are East-West Interfaces?
East-West interfaces enable horizontal coordination between peer OSS domains, orchestrators, and domain controllers at similar architectural levels. Unlike Northbound (upward) or Southbound (downward), East-West traffic flows horizontally between components that typically operate without strict hierarchy.
Why East-West Interfaces Matter
- Multi-domain orchestration – RAN controller talking to Transport controller coordinating 5G slice provisioning
- Service assurance correlation – Fault management sharing topology context with performance management
- Inventory federation – Resource inventory and service inventory exchanging updates and maintaining relationship consistency across OSS domains
- Closed-loop automation – Analytics/AIOps platform triggering orchestration workflows automatically based on SLA or anomaly conditions
- O-RAN architecture – SMO coordinating with Near-RT RIC, Non-RT RIC, and O-Cloud infrastructure
- 5G network slicing – Core, transport, and RAN domains coordinating slice lifecycle management
Why East-West Integration Is Challenging
- Different vendors use different orchestration models, APIs, and data models
- Cross-domain synchronization requires consistent inventory, topology, and relationship views
- Real-time orchestration introduces timing, latency, and dependency challenges
- Fault isolation becomes difficult across distributed domains – root cause may span multiple orchestrators
- Standardization across OSS domains (TMF, MEF, O-RAN, 3GPP) is still evolving and sometimes overlapping
- Security, authentication, and authorization must span across domains consistently
Common East-West Integration Scenarios
1. Assurance & Orchestration Integration
Service Assurance detects SLA degradation → East-West API calls Orchestration → Orchestration triggers re-routing or scaling → Assurance verifies resolution.
2. Inventory Federation
Resource inventory (physical devices) and service inventory (logical services) exchange updates and maintain relationship consistency across OSS domains.
3. O-RAN SMO & RIC Integration
SMO (Service Management and Orchestration) coordinates with Near-RT RIC (RAN Intelligent Controller) via East-West style interfaces for xApp/rApp lifecycle management and policy updates. O-RAN interfaces such as A1, O1, and O2 enable coordination between orchestration, management, RIC, and cloud domains.
4. Cross-Domain 5G Slice Orchestration
RAN domain orchestrator, Transport domain orchestrator, and Core domain orchestrator coordinate via East-West interfaces to provision end-to-end network slices.
Common East-West Protocols & Technologies
| Technology | Use Case | Example |
|---|---|---|
| REST APIs | Synchronous orchestration calls | Orchestrator querying inventory state |
| Kafka / Event Bus | Asynchronous event distribution | Fault events to multiple OSS consumers |
| gRPC | High-performance microservice communication | RIC ↔ SMO real-time policy exchange |
| TMF Open APIs | Standardized OSS domain integration | TMF639 (resource) ↔ TMF638 (service) synchronization |
| Message Queues (RabbitMQ, ActiveMQ) | Reliable message delivery | Workflow orchestration between OSS components |
Real-World Example: 5G Network Slice Orchestration
An operator provisions a low-latency 5G slice for autonomous vehicles:
- Northbound: BSS sends slice order to Service Orchestrator
- East-West (RAN ↔ Transport): RAN Orchestrator reserves spectrum; Transport Orchestrator allocates fibre capacity
- East-West (Core ↔ Transport): Core Orchestrator instantiates UPF; Transport ensures QoS guarantees
- East-West (Assurance ↔ Orchestration): Assurance/AIOps platform monitors slice KPIs and triggers scaling if needed
- Southbound: Configurations pushed to gNBs, routers, and core network functions
- Northbound: Slice activation confirmed back to BSS
Without East-West interfaces, each domain orchestration would operate in isolation, making end-to-end slicing impossible.
East-West vs North-South – Key Differences
| Aspect | North-South | East-West |
|---|---|---|
| Direction | Hierarchical (up/down) | Horizontal (peer or federated domains) |
| Purpose | Integration across architectural layers | Coordination within/across peer domains |
| Example | OSS → BSS (Northbound), OSS → Device (Southbound) | RAN Orchestrator ↔ Transport Orchestrator |
| Typical Protocols | SNMP, gNMI (South); TMF APIs, REST (North) | Kafka, gRPC, REST, message queues |
In multi-vendor environments, RAN Orchestrator (Nokia), Transport Orchestrator (Cisco), and Core Orchestrator (Ericsson) must interoperate via East-West interfaces. Standardization (TMF APIs, MEF LSO, O-RAN) is critical to avoid vendor lock-in.
Connection to BSS
BSS typically interacts with OSS via Northbound interfaces. However, East-West integration impacts BSS indirectly:
- Order to slice provisioning: BSS order triggers East-West orchestration across domains
- SLA assurance: Assurance platform (via East-West) collects performance from multiple domains for SLA validation
- Unified customer view: Service inventory (fed via East-West) provides BSS with service-to-resource relationship
Common Interview Questions
Q1. What is the difference between North-South and East-West interfaces?
North-South is hierarchical (different layers). East-West is horizontal coordination between peer or federated OSS domains at similar architectural levels.
Q2. Why are East-West interfaces critical for 5G network slicing?
Network slicing requires coordination across RAN, Transport, and Core domains. East-West interfaces enable domain orchestrators to exchange capacity, topology, and performance information.
Q3. What protocols are commonly used for East-West communication?
Kafka, gRPC, REST APIs, TMF Open APIs, and message queues (RabbitMQ). In cloud-native environments, service meshes and API gateways.
Q4. Why is East-West integration challenging?
Vendor diversity, cross-domain synchronization, real-time coordination requirements, fault isolation across domains, and evolving standards make East-West integration complex.
Q5. How does East-West relate to closed-loop automation?
Assurance/AIOps platform detects anomaly → East-West API calls Orchestration → Orchestration triggers remediation → Assurance verifies resolution. No human intervention.
Key Terms
Takeaways for You
- East-West interfaces enable horizontal coordination between peer or federated OSS domains and orchestration components.
- North-South = hierarchical (different layers). East-West = horizontal (peer domains, same or adjacent layers).
- Critical for: 5G slicing, O-RAN, multi-domain orchestration, closed-loop automation, inventory federation.
- Common protocols: Kafka, gRPC, REST APIs, TMF Open APIs, message queues, service meshes.
- Challenges include: vendor diversity, cross-domain synchronization, real-time coordination, fault isolation, and evolving standards.
- Without East-West: Domain orchestrators operate in isolation → end-to-end services impossible.
Recommended Next Learning Path