Backlog-Aware Queuing for Distributed Compute Function Routing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing edge computing networks face challenges in efficiently delivering network services without a centralized controller, particularly in determining function execution and data production locations, coordinating routing decisions, and adapting to time-varying demand rates, leading to inefficiencies and high operational costs.
Innovation Solution
A distributed queuing control framework with service discovery procedures for dynamic orchestration in data centric networks (DCNs) that enables local decision-making based on request backlog observations and service discovery, allowing nodes to dynamically schedule function execution and data forwarding, optimizing network throughput and latency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a centralized controller is used for network service delivery, then coordination and control are improved, but system complexity and operational costs increase
Solution Approach 1:
The patent extracts the centralized controller function and distributes control logic across individual network nodes. Each node autonomously makes routing decisions based on local queue backlog observations, eliminating the need for a centralized controller while maintaining coordinated service delivery through distributed queuing control mechanisms.
Solution Approach 2:
Network nodes perform self-control by monitoring their own queue backlogs and autonomously determining routing decisions. The system enables self-service through local nodes making control decisions based on observed backlogs, without requiring external centralized management, thus reducing system complexity while maintaining reliability.
2Device complexity
If static routing decisions are used, then control simplicity is improved, but adaptability to time-varying demand rates deteriorates
Solution Approach 1:
The patent implements dynamic routing decisions that adapt to time-varying demand rates through continuous monitoring of queue backlogs. Routing paths are dynamically adjusted based on real-time backlog observations, allowing the system to respond to changing network conditions while maintaining relatively simple control logic at each node.
Solution Approach 2:
The system employs feedback mechanisms where nodes observe queue backlogs and use this information to adjust routing decisions. This feedback loop enables adaptability to time-varying demand rates while keeping control simple, as nodes automatically respond to backlog changes without complex centralized coordination.
3Adaptability or versatility
If distributed control without centralized controller is implemented, then system scalability is improved, but coordination efficiency deteriorates
Solution Approach 1:
The patent segments the control function across individual network nodes, with each node independently monitoring local queue backlogs and making routing decisions. This segmentation enables system scalability by allowing nodes to operate autonomously, while coordination efficiency is maintained through direct local observations and immediate responses to backlog changes.
Solution Approach 2:
Queue backlog observations serve as an intermediary mechanism that enables coordination between distributed nodes. Nodes use backlog information as a shared reference to make routing decisions, facilitating efficient coordination without requiring a centralized controller, thus achieving both scalability and coordination efficiency.
4Reliability
If function execution locations are determined centrally, then service delivery reliability is improved, but operational costs increase
Solution Approach 1:
The patent extracts centralized function location determination and implements distributed decision-making where nodes autonomously select execution locations based on local queue backlog observations. This reduces operational costs by eliminating centralized control overhead while maintaining service delivery reliability through locally-optimal routing decisions.
Solution Approach 2:
Nodes perform self-service by independently determining function execution locations based on their own queue backlog conditions. This self-determination mechanism reduces operational costs associated with centralized management while ensuring reliable service delivery through locally-adapted routing decisions that respond to real-time network conditions.
Data Source
AI summary
In one embodiment, a node of a data centric network (DCN) may receive a first service request interest packet from another node of the DCN, the first service request interest packet indicating a set of functions to be performed on source data to implement a service. The node may determine that it can perform a particular function of the set of functions, and determine, based on a backlog information corresponding to the particular function, whether to commit to performing the particular function or to forward the service request interest packet to another node. The node may make the determination further based on service delivery information indicating, for each face of the node, a service delivery distance for implementing the set of functions.