Blockchain-Based Distributed Computing Verification System
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The challenge lies in efficiently verifying the accuracy of processing results in a distributed computing system where providers may not have strong incentives to ensure correctness, and complex systems like cloud servers often experience misconfigurations and hardware randomness, making it difficult to guarantee execution accuracy.
Innovation Solution
A blockchain-based system is employed, utilizing officer nodes, witness nodes, and enforcement nodes to verify processing results through fishing tasks, where officer nodes send tasks to witness nodes for confirmation, and enforcement nodes ensure validity by comparing results and taking corrective actions, with a decentralized and trustable mechanism.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If processing tasks are outsourced to distributed computing nodes, then computing power utilization is improved, but result verification accuracy deteriorates
Solution Approach 1:
The system performs preliminary actions by having officer nodes generate fishing tasks and store their correct results in the blockchain before actual processing occurs. This pre-computation and pre-storage of verification data enables later verification without re-executing the original computation, thus maintaining verification accuracy while enabling distributed processing.
Solution Approach 2:
Enforcement nodes act as intermediaries between officer nodes and processing nodes. They receive fishing tasks from officer nodes, distribute them to processing nodes, collect results, and verify correctness by comparing against stored results in the blockchain. This intermediary layer enables scalable verification without requiring direct communication between all nodes.
2Reliability
If fishing tasks are sent to multiple processing nodes for verification, then result reliability is improved, but system complexity increases
Solution Approach 1:
The verification system is segmented into distinct node types with specialized functions: officer nodes generate and store fishing tasks, enforcement nodes distribute tasks and verify results, and processing nodes execute computations. This segmentation allows each node type to perform its function independently, improving reliability through distribution while managing complexity through functional specialization.
Solution Approach 2:
Processing nodes serve multiple functions: they can act as standard processing nodes executing user tasks, as witness nodes providing high-reliability processing for fishing tasks, and as targets for verification. This multi-functionality allows the same infrastructure to support both production workloads and verification activities, reducing overall system complexity.
3Reliability
If officer nodes store fishing tasks and results in blockchain, then verification trustability is improved, but storage requirements increase
Solution Approach 1:
Enforcement nodes maintain copies of fishing tasks and results from the blockchain for local verification. Instead of requiring all verification to reference the original blockchain storage, enforcement nodes can work with local copies, reducing the need for continuous blockchain access and distributing storage requirements across the network.
Data Source
AI summary
Methods and apparatus for detecting that a processing node, in a network including a plurality of processing nodes, is reporting invalid results and for taking corrective actions in response to the detection are described.