MAPE-K Self-Healing Framework Achieves 93.2% Recovery Rate
A novel modular self-healing framework, utilizing the MAPE-K model, has reached a recovery success rate of 93.2% for web applications. This system incorporates an AutoFix-inspired method for adaptive fault recovery. Through a design and development research methodology, the framework underwent testing with controlled fault injections across twenty different runtime failure scenarios, such as memory leaks, service crashes, and database disconnections. It achieved a mean fault detection F1-score of 90.7%. The AutoFix component improved average recovery time by 56.2%, resulting in an average of 3.92 seconds. Throughout the recovery process, system throughput remained stable. This research tackles the significant issue of maintaining reliability in complex, dynamic web environments.
Key facts
- Framework based on MAPE-K model with AutoFix-inspired mechanism
- Tested via controlled fault injection across 20 runtime failure scenarios
- Mean fault detection F1-score of 90.7%
- Recovery success rate of 93.2%
- AutoFix module reduced average TTR by 56.2%
- Average recovery time of 3.92 seconds
- System throughput maintained during recovery
- Addresses reliability challenges in complex web applications
Entities
—