Distributed Service Function Orchestration Across Network Domains
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional distributed routing protocols in multi-network domains fail to efficiently manage service function chaining due to high computation complexity and signaling overhead, particularly when considering the availability of service functions and virtual machines across geographically distributed data centers.
Innovation Solution
A distributed resource orchestration framework using vertex-centric distributed processing that enables different vertices to exchange information iteratively to identify qualified service function chain solutions without requiring a global view or database, allowing for end-to-end service requests across multiple domains.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If traditional distributed routing protocols are used to compute network paths, then routing functionality is provided, but service function availability and virtual machine availability at individual network nodes cannot be considered
Solution Approach 1:
The patent introduces a service function chain orchestrator as an intermediary component that bridges traditional routing protocols and service function chaining requirements. This orchestrator maintains a global view of service functions and virtual machines across multiple network domains, enabling path computation that considers service function availability without requiring modification of existing routing protocols. The orchestrator acts as a mediator between the routing infrastructure and the service function deployment, resolving the contradiction by adding orchestration complexity only where needed while preserving simplicity in the underlying routing system.
Solution Approach 2:
The patent segments the orchestration functionality into domain-specific orchestrators and a global orchestrator. Each domain orchestrator manages service functions within its own network domain, maintaining local state information. The global orchestrator coordinates across domains, maintaining a global view of service function chains. This segmentation allows the system to consider service function availability across domains while distributing the computational complexity, preventing any single component from becoming a bottleneck or single point of failure.
2Adaptability or versatility
If a hybrid architecture with geographically distributed orchestrators replicating a global view is used, then service function chaining across domains is enabled, but managing the global network state and maintaining confidentiality of various network domains becomes challenging
Solution Approach 1:
The patent implements segmentation by creating domain-specific orchestrators that maintain local state information about service functions within their respective network domains. Each domain orchestrator keeps confidential information about its domain's service functions and virtual machines locally, rather than replicating the entire global view across all orchestrators. The global orchestrator maintains a high-level view sufficient for coordination but does not expose detailed confidential information about individual domains. This segmentation approach enables multi-domain service orchestration while preserving domain confidentiality by ensuring that sensitive information remains localized.
Solution Approach 2:
The patent applies local quality by allowing each domain orchestrator to maintain its own local state information with domain-specific confidentiality requirements. Each orchestrator has tailored knowledge about its local domain's service functions, virtual machines, and network characteristics. This local quality approach enables the system to respect domain boundaries and confidentiality requirements while still achieving global coordination through controlled information exchange between orchestrators.
3Productivity
If path-computation element based multi-domain heuristics are used to map virtual optical network requests, then inter-domain path computation is achieved, but significant signaling overhead occurs due to computing all possible inter-domain paths
Solution Approach 1:
The patent applies preliminary action by having domain orchestrators pre-compute and cache feasible paths within their own domains before global orchestration is needed. When a service function chain request spans multiple domains, the global orchestrator can leverage these pre-computed local paths rather than computing all possible inter-domain paths from scratch. This preliminary preparation significantly reduces the signaling overhead and computational burden during actual service deployment, as the system only needs to coordinate between pre-validated local paths rather than exploring the entire solution space.
4Productivity
If centralized orchestrators maintain hierarchical topology for all domains in multi-network domains, then service function chain requests can be mapped, but scalability and computational complexity become problematic on large-scale networks
Solution Approach 1:
The patent resolves the scalability problem by segmenting the centralized orchestration into distributed domain orchestrators, each managing its own network domain. Instead of one centralized orchestrator maintaining hierarchical topology for all domains, multiple domain orchestrators each maintain local topology information about their respective domains. This segmentation distributes the computational complexity and memory requirements across multiple independent components, enabling the system to scale to large networks with many domains without overwhelming any single orchestrator.
Solution Approach 2:
The patent transitions from a vertical hierarchical orchestration model to a horizontal distributed model. Instead of organizing orchestration in a single hierarchical dimension with a root orchestrator at the top, the system organizes orchestrators horizontally across domain boundaries, with each orchestrator operating at its own level. This dimensional change allows the system to maintain global coordination capabilities while distributing the computational burden across multiple peers, improving scalability and reducing the complexity burden on any single component.
Data Source
AI summary
In one or more embodiments, one or more systems, methods, and/or processes may determine one or more auxiliary edges that bypass at least one vertex of vertices that represent physical nodes of a network domain; evaluate at least one edge, that includes the one or more auxiliary edges and that interconnect the vertices, to evaluate a portion of the vertices that excludes the at least one vertex that was bypassed to identify at least one vertex that is associated with at least one service function of a service function chain request specifying service functions to be performed via at least a portion of physical nodes of network domains; and configure a first physical node of the physical nodes of the network domain and associated with the at least one vertex that is associated with the at least one service function to process data via the at least one service function.