Reliable Computing

A computer that does its job, even when something goes wrong. Reliable computing means a system works as expected. We deal with everything that threatens it: faults, errors and failures, and how to overcome them.

What we work on

• Detecting, containing and correcting errors before they do harm
• Resilient systems, for example for industrial communication
• Assessing the reliability of electronics and computing systems

Checking reliability early in the design

We build system models of processors and accelerators in SystemC and use them to study the trustworthiness of a design long before there is hardware:

• how faults behave and propagate
• which security level a design reaches
• whether the design is safe

Reliability in electronics

• reliable accelerators based on RRAM, an emerging resistive memory
• reliable cryptographic accelerators
• analysing and evaluating the reliability of systems-on-chip
• HW/SW co-verification for reliable IP blocks and integrated circuits

From our research: Trustworthy Integrated Circuits: From Safety to Security and Beyond (IEEE Access, 2024), Holistic Framework for Evaluating the Trustworthiness of Integrated Circuits and Secure Software/Hardware Hybrid In-Field Testing for System-on-Chip (both VLSI-SoC 2024).

The group is led by Dr.-Ing. Saleh Mulhem.