Considerations on the Design of Resilient System-in-Package Based on Machine Learning and On-Chip Lifecycle Management

F. Vargas, M. Andjelkovic (IHP Frankfurt (Oder), Germany), M. Taheri (Brandenburg Univ. of Techn. Cottbus–Senftenberg, Germany and Taltech, Estonia)

This tutorial addresses the benefits one can take from deploying on-chip lifecycle management (OCLM) to design resilient Systems-in-Package (SiP). It overviews the substantial advancement that the design of chiplet-based AI systems has achieved in the recent years. The benefits of using OLCM are based on the massive deployment of on-chip cross-layer, collaborative sensors that are placed in strategic points of the chiplets and interposer, a data wrapper interface & management hub that is responsible to process and store the collected data from the sensors in a way to create an in-mission history of events, and on-chip analytics engine. The sensors can be placed in the chiplets and on the interposer. Moreover, the sensors can monitor process-voltage-temperature (PVT) at the silicon level, as well as monitor data transactions at the RTL and operating system level. IEEE and IEC standards can be used to support data monitoring and collection processes. The analytics engine can be based on simple/straightforward data structures or on advanced Machine Learning/Artificial Intelligence (ML/AI) algorithms.