Machine learning, big-data and internet-of-things are some of the emerging computing applications which are extremely demanding in terms of storage, energy and performance. While conventional von Neumann architectures are facing significant challenges to cope with such demands, computing-in-memory architectures in general and neuro-inspired architectures, in particular, represent a promising solution to overcome these limitations.
This Master/Bachelor Thesis aims to develop a virtual prototype to perform a system exploration concerning parameters like power consumption and performance. The virtual prototype will be implemented in C++ together with SystemC/TLM.
Among others, these steps must be done:
- Literature research about neuromorphic accelerators
- Implement a virtual prototype of a neuromorphic accelerator
- Include this virtual prototype into our already existing virtual platform
- Implement power tracing to estimate the power consumption
- Evaluation and verification by means of simulation