Service Function Network Layer for Scalable Chain Management
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current approaches to managing service function chains in computer networks are inefficient and inflexible, requiring complex configuration procedures, excessive physical resources, and reconfiguration of network switches for modifications, which limits scalability and compatibility.
Innovation Solution
A self-contained service function network layer using chain tables and APIs to manage and route packets through a substrate network, allowing for dynamic deployment, reconfiguration, and modification of service function chains without reconfiguring underlying switches.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If SDN routing rules are used to dynamically deploy logical chains, then network flexibility and service deployment capability are improved, but the aggregate size of routing tables grows exponentially and memory requirements increase
Solution Approach 1:
The patent segments the routing functionality by separating service chain routing from traditional network routing. Instead of maintaining one large exponential routing table, the system divides routing into: (1) standard network routing for substrate network topology, and (2) service chain routing for logical chain deployment. This segmentation prevents the exponential growth of a single routing table while maintaining both network and service routing capabilities.
Solution Approach 2:
The patent introduces a new dimensional layer for service chain routing that operates independently from traditional network routing. Service chains are routed in a logical dimension above the physical network layer, allowing service-specific routing rules to be applied without expanding the underlying network routing tables. This dimensional separation enables service deployment flexibility without proportionally increasing routing table sizes.
2Reliability
If physical resources are allocated for each network function, then service reliability and performance are improved, but the physical resources required increase rapidly with large scale services
Solution Approach 1:
The patent implements universal virtualized network function instances that can serve multiple service chains simultaneously. A single physical resource running a virtualized network function (e.g., firewall, NAT, intrusion detection) can be dynamically assigned to different service chains based on demand. This multi-functionality approach maintains service reliability through dedicated virtual instances while dramatically reducing the total physical resources required compared to having separate physical hardware for each service instance.
3Adaptability or versatility
If network switches are reconfigured when logical chains are modified, then service chain flexibility is improved, but the complexity of network management increases and service interruption occurs
Solution Approach 1:
The patent extracts service chain routing logic from the network switches and consolidates it in the SDN controller. When service chains need to be modified, added, or removed, only the controller's service chain routing table needs to be updated. The network switches continue to operate with their existing network routing configurations, eliminating the need for switch reconfiguration. This extraction reduces network management complexity and prevents service interruptions while maintaining service chain flexibility.
4Quantity of substance
If ternary content-addressable memory (TCAM) is used for SDN logical chains, then routing rule storage capability is improved, but the cost increases significantly and design limitations exist
Solution Approach 1:
The patent uses standard RAM-based data structures to create virtual copies of routing functionality instead of relying on expensive TCAM hardware. Service chain routing rules are stored as software data structures in conventional memory, which can be dynamically created, modified, and deleted without hardware reconfiguration. This copying approach provides sufficient routing rule storage capability while eliminating the high cost and design limitations of TCAM hardware.
Data Source
AI summary
Embodiments of the present disclosure describe a self-contained service function network layer between a chain (e.g., a logical chain or service function chain) and a substrate network. Embodiments of the present disclosure further provide techniques for constructing logical chains for service function networks using chain tables, organizing chain tables using application program interfaces (APIs), deploying service function networks to substrate networks, routing packets through a service function network and a substrate network, and inserting, deleting, re-routing, moving and substituting service functions in logical chains.