一种DRAM内存颗粒的测试方法、设备及介质
By employing spatiotemporal graph evolution and failure coupling inference methods, the problem of continuously depicting error evolution relationships in existing DRAM testing is solved, thereby improving the accuracy of constructing health profiles and determining faults for DRAM memory chips.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- SHENZHEN COMOS INTELLIGENT TECHNOLOGY CO LTD
- Filing Date
- 2026-04-29
- Publication Date
- 2026-07-17
AI Technical Summary
Existing DRAM testing methods lack the ability to characterize the continuous evolution of errors in both time and space dimensions when faced with complex failure behaviors that exhibit repeated triggering characteristics and time-evolution patterns.
The spatiotemporal graph evolution method is used to perform structured evolutionary analysis on the abnormal log set, generate time-related sequences, extract repeated triggering features, and combine the failure distribution set with repeated triggering features through the failure coupling inference method to generate a granular health profile, and conduct targeted verification tests.
It improves the ability to analyze complex failure behaviors and the accuracy of fault determination, and can continuously characterize the evolution of errors in the time and space dimensions, guiding targeted verification testing and fault classification.
Smart Images
Figure CN122112554B_ABST