Non-Redundant Fault Processing TRIZ Cases
- Intelligent Feature Controller for Network Failure Containment
An AI/ML controller maps feature dependencies, disables affected features, and limits network-wide collateral damage after failures.
- Machine-Learned Fix Verification Procedures from Historical Defects
Historical defect data is matched to current problem analysis, ranking verification procedures to reduce developer-agent back-and-forth and downtime.
- ESD Detection in Data Storage Devices for Clock Sync Protection
An ESD antenna sensor disables sensitive signals and freezes the reference clock to prevent sync loss during discharge events.
- Virtualized Fault Handler Reconfiguration Without System Shutdown
Virtualized fault handlers let engineers update recovery actions and clear event status without suspending applications or rebooting the system.
- Intra-Block Redundancy Checks for Distributed Storage Recovery
Reconstruct distributed dataset sub-portions, detect mismatches, and replace differing data to support intra-block error correction.
- Storage I/O Path Anomaly Detection with Task-Sequence Learning
Parsed storage traces are classified with machine learning to separate anomalous I/O task sequences from normal operations and reduce false alerts.
- Multi-Tier Network Health Checks for Grey-Failure Rerouting
Tiered health checks detect partial grey failures and reroute communication flows around affected network elements to maintain continuity.
- Recovery Module Monitoring for Data Migration Failure Events
During logical-device data migration, the recovery module detects failure events and initiates remediation to complete recovery.