Service Chaining via SPID Encapsulation in Provider Networks
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing network technologies face challenges in efficiently processing data packets through multiple services in a sequential order within a service provider network, requiring a flexible mechanism to steer data packets to the correct service nodes according to service chaining policies.
Innovation Solution
A method involving a flexible map and encapsulation mechanism that uses Service Path Identifiers (SPIDs) and Endpoint Identifiers (EIDs) and Routing Locators (RLOCs) to sequentially route data packets through service nodes, enabling dynamic service placement and scalability.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If data packets are routed through multiple services in sequential order, then service processing completeness is improved, but network complexity and routing difficulty increase
Solution Approach 1:
The patent segments the service chain into discrete service nodes, each with unique EIDs, and separates routing functions into ingress nodes, egress nodes, and service nodes. This segmentation allows complex service processing to be broken down into manageable sequential steps while maintaining overall system organization and reducing routing complexity through clear role definition.
Solution Approach 2:
The patent introduces intermediary components including the service chain policy database that stores and manages service chaining policies, and the ingress/egress nodes that act as mediators between external networks and internal service nodes. These intermediaries abstract the complexity of service chaining from end systems and provide centralized policy management.
2Reliability
If service chaining policies are enforced for all data packet flows, then service quality is improved, but processing overhead and time increase
Solution Approach 1:
The patent implements preliminary action by pre-configuring service chaining policies in the service chain policy database before data packet flows arrive. The ingress nodes retrieve pre-determined service chains from the database, avoiding real-time policy computation and reducing processing time for each data packet while ensuring service quality through enforced policy compliance.
3Adaptability or versatility
If new services are added to the network, then service variety is improved, but scalability and configuration complexity worsen
Solution Approach 1:
The patent implements universality through the service chain policy database that provides a unified mechanism for managing all service chaining policies regardless of the specific services involved. The standardized EID/RLOC structure and encapsulation mechanism work uniformly across different service types, allowing new services to be integrated without changing the fundamental routing architecture or increasing configuration complexity.
Solution Approach 2:
The patent enables easy addition of new services through parameter changes in the service chain policy database. New services are integrated by assigning them unique EIDs and configuring their service chaining policies, without requiring changes to the underlying network infrastructure or routing protocols. This allows service variety to expand while maintaining consistent configuration procedures.
4Adaptability or versatility
If encapsulation mechanisms are used for service node communication, then routing flexibility is improved, but packet processing overhead increases
Solution Approach 1:
The patent uses copying in the form of encapsulation where the original data packet is wrapped with additional header information containing EID and RLOC fields. This copying approach preserves the original packet while adding routing information, allowing flexible service node routing without modifying the core data payload. The encapsulated packets can be processed in parallel and the original packet integrity is maintained throughout the service chain.
Data Source
AI summary
A system and method for chaining one or more services in a service provider network. A service chaining policy and associated Service Path Identifier (SPID) are determined at an ingress node with respect to a particular data packet flow. If the service chaining policy involves one or more service nodes to be traversed by the data packet flow, each service node's EIDs and RLOCs are determined. A sequential data exchange process with the service nodes is effectuated using encapsulation of data packets based on the EIDs and RLOCs for obtaining services in accordance with the order of services set forth in the chaining policy.


