Non-Redundant Fault Processing TRIZ Cases
- Automatic Error Fixing for High-Availability Applications
A wrapper applies fixer routines to input arguments to produce new arguments and uses the resulting output when acceptable.
- CPU Machine Check Exception Timestamping for Root Cause Analysis
Storing occurrence timestamps in machine check banks orders errors chronologically, resolving complex log interpretation bottlenecks.
- Memory Controller Wear Leveling via Bit Order Reversal
A memory controller changes bit order during write operations to distribute wear evenly across non-volatile memory cells.
- Intelligent Quality Accelerator for Resource Exchange Validation
An intelligent quality accelerator system stratifies sampling of resource exchanges using a machine learning loop to confirm correct processing.
- Autonomous Instance Recovery in Distributed Processing Systems
Parallel processing instances autonomously recover from failures by storing and retrieving their own status information, eliminating standby system costs.
- Shared Memory Error Recovery for Multiprocessor Partitions
Removes requests from affected units in shared memory logic to prevent disruption of non-failing partitions during multiprocessor error recovery.
- Error Detector Block Placement for Soft Error Recovery
Compiler embeds error detector blocks after critical instructions to detect soft errors, eliminating vulnerable intervals and reducing performance overhead.