Multi-Domain Bandwidth Reservation With Distributed Path Computation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing service orchestration systems face high workload and infrastructure costs due to managing extensive network resource states across multiple domains, necessitating a more efficient and lightweight approach for end-to-end service orchestration and bandwidth reservation.
Innovation Solution
A method and system that decomposes end-to-end services into subservices, distributing path computation across network domain controllers, maintaining a lightweight orchestrator database with real-time resource information, and using a database schema for time-aware management of network resources and bandwidth-on-demand reservations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the service orchestration system manages all network resource states through extensive databases, then the E2E service can be accurately tracked and managed, but the system workload and infrastructure cost increase significantly
Solution Approach 1:
The patent divides the network management function into segments: the orchestrator handles only high-level E2E service orchestration and receives resource state information from domain controllers, while individual domain controllers manage their own network resource states. This segmentation reduces the orchestrator's database management workload while maintaining accurate E2E service tracking through distributed information collection.
Solution Approach 2:
Domain controllers act as intermediaries between the orchestrator and individual network domains. They collect and maintain detailed resource state information for their respective domains, then provide summarized information to the orchestrator. This intermediary approach allows the orchestrator to maintain accurate E2E service knowledge without directly managing all network resource states.
2Loss of information
If the service orchestration system maintains extensive databases for all network resources, then complete resource state information is available, but the infrastructure cost increases
Solution Approach 1:
The patent segments database storage responsibilities: domain controllers maintain detailed resource state databases for their specific domains, while the orchestrator maintains only high-level E2E service information and aggregated resource state summaries. This segmentation dramatically reduces the total database storage requirement while preserving complete network resource state information through distributed storage.
Solution Approach 2:
The patent extracts detailed resource state management from the orchestrator and assigns it to domain controllers. The orchestrator extracts only the essential E2E service orchestration information it needs, while domain controllers retain the detailed resource state data. This extraction reduces the orchestrator's infrastructure cost while maintaining information availability.
3Ease of operation
If path computation is performed centrally by the orchestrator, then end-to-end service routing is coordinated, but the computational workload on the orchestrator increases
Solution Approach 1:
The patent segments path computation into two parts: domain controllers perform local path computation for their respective domains based on requests from the orchestrator, while the orchestrator performs only high-level E2E service path coordination by aggregating results from domain controllers. This segmentation reduces the orchestrator's computational workload while maintaining effective E2E service coordination.
Solution Approach 2:
The patent transitions from centralized monolithic path computation to a distributed hierarchical approach where domain controllers perform detailed local computation and the orchestrator performs coordination at a higher level. This dimensional change in computation distribution reduces orchestrator processing requirements while maintaining E2E service coordination capability.
Data Source
AI summary
Disclosed herein are methods and systems for E2E service decomposition to orchestrate and provision services across multiple network domains (a plurality of network domains), and time-aware calendarization. The methods and systems provide time-aware E2E service reservation (e.g., bandwidth allocation) and provisioning across the multiple network domains based on a method and system model that manages the multiple network domains. and methods for verifying and reserving bandwidth and other E2E service reservations.