Self-healing web apps achieve 93% recovery success with MAPE-K framework
New framework detects faults in 3.9 seconds and cuts recovery time by 56%.
A new research paper from Sales Aribe Jr. and Rov Japheth Oracion introduces a self-healing framework for web applications that uses the MAPE-K (Monitor-Analyze-Plan-Execute over a shared Knowledge base) model, enhanced with an AutoFix-inspired recovery module. The team tested the system across 20 different runtime failure scenarios, including service crashes, memory leaks, and database disconnections, using controlled fault injection experiments. Results showed a mean fault detection F1-score of 90.7% and a recovery success rate of 93.2%.
The AutoFix module reduced average time-to-recovery (TTR) by 56.2%, bringing it down to just 3.92 seconds. During fault conditions, system throughput remained between 88% and 95% of normal levels, with only a 3.1% increase in response time. The iterative feedback mechanism improved recovery efficiency by 18.6% over multiple cycles. While the framework currently relies on predefined recovery strategies, the integration of learning-oriented feedback lays groundwork for more autonomous self-healing systems in the future.
- Framework achieved 90.7% F1-score in fault detection and 93.2% recovery success across 20 failure scenarios
- Time-to-recovery reduced by 56.2% to an average of 3.92 seconds using the AutoFix module
- System throughput maintained at 88–95% during faults with only 3.1% added latency
Why It Matters
Practical self-healing for web apps means fewer outages, faster fixes, and resilient operations for critical online services.