.

M. Sc. Nidhi Anantharajaiah

  • 24.04.2024
  • Adaptive and Distributed Networks-on-Chip for Mixed Criticality Systems
  • Group: Prof. Becker
  • Corrector: Prof. Dr. sc. techn. Andreas Herkersdorf (Technische Universität München)

Research

Adaptive networks-on-chip for mixed-criticality systems

To reduce resource consumption, multiple applications with different criticality are increasingly being executed on the same system-on-chip (SoC) platform. These platforms are moving towards multi/many-core architectures and can use networks-on-chip (NoC) for communication. As the number of cores on such platforms increases, the complexity of interconnects increases and becomes a limiting factor for performance. The performance of such systems can be improved if the NoC is able to adapt to the requirements of different applications at runtime.

Routing algorithms for non-mesh based topologies in a Networks-on-Chip

Mesh topology is often implemented in Networks-on-Chip (NoC) because it is scalable, easy to implement and uniform. With the increasing complexity of on-chip connections for large multi/many-core systems, the use of mesh topology can become inefficient due to the high number of hops. Hierarchical and multi-layer topologies as well as irregular topologies are being explored to improve the overall performance of the system. Routing in such NoCs can become a challenge, which is why studies on routing algorithms for regular and irregular topologies are conducted here.

Open student projects

title type