Arithmetic Unit Segmentation for Fault-Tolerant Calculation Efficiency
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In fault-tolerant computers, existing synchronization schemes like lock step fail to effectively utilize calculation resources when multiple cores are involved, leading to inefficient resource allocation and prolonged standby times for cores that have not executed output requests.
Innovation Solution
An information processing system is designed with a precedence system and a delay system, where the precedence system stops calculation and transfers data to the delay system, allowing non-output cores to continue processing, and the delay system reproduces the calculation and compares results, enabling effective resource utilization by allowing other cores to execute secondary calculations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If all cores stop calculation when an output request is generated to ensure synchronization and fault tolerance, then calculation reliability is maintained, but calculation resource utilization deteriorates
Solution Approach 1:
The patent segments the cores into two groups: cores that have executed output requests (first group) and cores that have not executed output requests (second group). This segmentation allows differential treatment where only the first group stops calculation for synchronization while the second group continues calculating, thus maintaining reliability for output-critical cores while improving overall resource utilization through continued computation in the second group.
2Measurement precision
If cores wait for comparison results from the delay system before resuming calculation, then calculation accuracy is ensured, but calculation time increases
Solution Approach 1:
The patent applies preliminary action by having the delay system perform calculation reproduction and comparison in advance before the precedence system needs to resume. The delay system calculates using the same initial state and execution commands, then compares results beforehand. This allows the precedence system to quickly resume calculation once accuracy is confirmed, reducing overall calculation time while maintaining accuracy through pre-performed verification.
Data Source
AI summary
An information processing system includes: a first system that includes a group of arithmetic units, a controller, and an external device; and a second system configured to execute calculation which is the same as calculation executed in the first system and compare calculation results to each other, wherein the controller is configured to: stop a plurality of arithmetic units when it is detected that an output request to the external device is output from one or more arithmetic units among the plurality of arithmetic units that execute first calculation in the group of arithmetic units, the plurality of arithmetic units including one or more arithmetic units that does not output the output request, transmit first comparison target data including a value output in response to the output request to the second system, and instruct the stopped one or more arithmetic units to execute second calculation.