Because Moore’s Law, i.e. that the number of transistors on a chip is doubling roughly every two years, is slowly dying1, it is worthwhile to look into new technologies that complement or replace traditional CMOS-based chip designs, in order to keep progressing. Promising candidates are memristive technologies. To put their operation principle simple: They are based on materials with electrically controllable resistance.
This principle can be exploited in multiple ways:
- They can be used to store data. One remarkable property of memristors as storage elements is, that they will keep their state without an attached voltage and therefore lend themselves as basis of non-volatile memory (NVM)2.
- They can also be used as computational devices. A lot of research has been done on memristors facilitating neuromorphic computation, because they seem to be able to mimic synapses reasonably well3. Also they can be used to build up logic gates4.
Because of these properties we are exploring the possibilities offered by memristors on the circuit, chip and system level to generate fast and energy efficient processors.
- T. N. Theis and H. S. P. Wong, “The End of Moore’s Law: A New Beginning for Information Technology,” Computing in Science Engineering, vol. 19, no. 2, pp. 41–50, Mar. 2017.↩
- Y. Ho, G. M. Huang, and P. Li, “Nonvolatile Memristor Memory: Device Characteristics and Design Implications,” in Proceedings of the 2009 International Conference on Computer-Aided Design, New York, NY, USA, 2009, pp. 485–490.↩
- B. Linares-Barranco, T. Serrano-Gotarredona, L. A. Camuñas-Mesa, J. A. Perez-Carrasco, C. Zamarreño-Ramos, and T. Masquelier, “On Spike-Timing-Dependent-Plasticity, Memristive Devices, and Building a Self-Learning Visual Cortex,” Front. Neurosci., vol. 5, 2011.↩
- L. Xie, H. A. D. Nguyen, M. Taouil, S. Hamdioui, and K. Bertels, “Boolean logic gate exploration for memristor crossbar,” in 2016 International Conference on Design and Technology of Integrated Systems in Nanoscale Era (DTIS), 2016, pp. 1–6.↩