Fault-Tolerant Function Placement in Edge Computing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for function placement in edge computing environments often prioritize latency over fault tolerance, leading to systems that are vulnerable to failures and errors, especially when computational tasks are distributed across multiple placement locations.
Innovation Solution
The implementation of an application placement service that uses a directed acyclic graph (DAG) to model application flows, predicts latency and failure rates, and optimizes function placement across multiple computing tiers to achieve both low latency and low error rates, employing machine learning and automated provisioning to streamline deployment.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Speed
If functions are placed at edge locations to reduce latency, then speed is improved, but reliability deteriorates due to higher failure rates at edge locations
Solution Approach 1:
The patent introduces a new dimension to function placement by considering multiple placement locations across different tiers (edge, regional, global) rather than single-location placement. This multi-dimensional placement strategy allows the system to optimize for both latency and reliability simultaneously by distributing functions across the hierarchy.
Solution Approach 2:
The system performs preliminary analysis of failure rates and latencies for different placement locations before making placement decisions. By pre-calculating these metrics and using them to guide placement, the system proactively avoids placements that would compromise reliability while maintaining low latency.
2Reliability
If functions are distributed across multiple placement locations to improve reliability, then reliability is improved, but device complexity increases
Solution Approach 1:
The patent transforms the complex multi-factor optimization problem into a simpler decision process by changing parameters to discrete tiers with predefined characteristics. Each tier has predetermined failure rate and latency profiles, allowing the system to make placement decisions based on tier selection rather than evaluating every possible location combination.
Solution Approach 2:
The tier-based placement system serves multiple functions simultaneously: it optimizes for latency, minimizes failure rates, simplifies the placement decision process, and provides a scalable framework that can adapt to different application requirements without increasing complexity.
3Measurement precision
If machine learning models are used to predict latency and failure rates, then measurement precision is improved, but device complexity increases
Solution Approach 1:
The system employs machine learning models that automatically learn from historical data and improve predictions over time without requiring manual intervention or complex configuration. The models self-adjust to capture patterns in latency and failure rate variations, providing high measurement precision while keeping the operational complexity manageable.
Data Source
AI summary
Techniques for fault-tolerant function placement across multiple computing placement locations are described. An application placement service obtains a latency-based placement of functions of an application and utilizes resource information associated with the placement locations to determine error rates for flows of the placement. For flows having an error rate falling outside of an allowable range, the application placement service can modify the placement to improve the error rate, e.g., by adding additional redundancy for functions in higher-error placement locations within those placement location tiers, and/or by adding additional redundancy for functions in higher-error placement locations within different placement location tiers.