Project Details
MOTARO: A modular approach for test and repair of embedded non-volatile memories.
Applicant
Professor Dr.-Ing. Rolf Kraemer
Subject Area
Computer Architecture, Embedded and Massively Parallel Systems
Electronic Semiconductors, Components and Circuits, Integrated Systems, Sensor Technology, Theoretical Electrical Engineering
Electronic Semiconductors, Components and Circuits, Integrated Systems, Sensor Technology, Theoretical Electrical Engineering
Term
from 2013 to 2018
Project identifier
Deutsche Forschungsgemeinschaft (DFG) - Project number 242177335
Existing techniques used for mitigating faults in scaled emerging non-volatile memory systems are unsatisfactory due to overheads imposed on the systems performance and behavior. In our project we will address and solve disadvantages of current fault-tolerant techniques and challenges of emerging non-volatile memories.The main idea behind the project is based on unprecedented combination of error correcting codes and built-in self-test and self-repair mechanisms. According to our idea the challenge of online memory test and repair will be divided into well-specialized tasks which can be performed on memory block- and word-level. By doing so, dependability of memories will be increased significantly and systems overheads associated with memory fault management will be reduced. We will show that our techniques can be implemented in form of an integrated controller which makes it an ideal and universal solution for different types of existing and emerging non-volatile memories.
DFG Programme
Research Grants