Transactional Graph Computation Error Handling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing graph-based computation systems face inefficiencies in processing transactions due to resource management and error handling limitations, particularly in concurrent transaction processing and handling errors within complex computation graphs.
Innovation Solution
The system implements a method for processing transactions using computation graphs by associating transactions with instances of computation graphs, enabling concurrent processing and error handling through reusable computation elements and error-handling graph components, which redirect processing to handle errors and maintain resource efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If multiple transactions are processed concurrently using graph-based computations, then productivity is improved, but system complexity increases due to resource management and error handling requirements
Solution Approach 1:
The system segments transactions into discrete work elements that can be independently processed through graph-based computation. Each work element represents a unit of work that can be assigned to computation elements, allowing concurrent processing while maintaining manageable complexity through modular organization of transaction processing
Solution Approach 2:
The patent implements reusable computation elements that can be shared across multiple graph instances and transactions. These computation elements represent universal processing units that can handle different types of work elements, reducing the need for dedicated resources for each transaction and thereby reducing overall system complexity while maintaining high productivity
2Reliability
If error handling is implemented in graph-based computation, then reliability is improved, but device complexity increases due to additional error-handling components
Solution Approach 1:
The patent introduces specialized error-handling graph components that act as intermediaries between computation elements and the overall system. These components intercept and manage errors without requiring changes to the core computation logic, isolating error handling complexity from the main computation graphs while improving reliability
Solution Approach 2:
The system creates copies of computation graphs for different transactions, allowing errors in one graph instance to be isolated and handled independently without affecting other instances. This copying approach enables reliable concurrent processing by ensuring that error handling in one transaction does not propagate to others
3Loss of energy
If reusable computation elements are used across multiple graph instances, then resource efficiency is improved, but the difficulty of detecting and measuring errors increases
Solution Approach 1:
The patent implements feedback mechanisms where error-handling components monitor and report errors back to the system. This feedback loop enables the system to track errors in reusable computation elements across multiple graph instances, maintaining error detectability despite the sharing of computation resources among multiple transactions
Data Source
AI summary
Processing transactions using graph-based computations includes determining that at least one of a plurality of graph elements of a computation graph of a set of one or more computation graphs includes a computation to be performed for a given transaction, associating the given transaction with an instance of the computation graph that includes reusable computation elements associated with respective graph elements, and executing the graph to perform the computation.