Why do we need anything like Neuromorphic computing? Isn’t current computing sufficient for us?
Yes, it is, for now. But we need neuromorphic computing. Mainly due to energy efficiency, possibility of down-scaling to improve data integration density and in-memory computing to minimize loss of time and energy in transfer of data between memory and processor units. IoE and IoT will require storing and processing of unprecedented amount of data. This will have a huge energy burden. Moreover the current CMOS technology will soon be unable to go for further miniaturization posing a serious challenge on increasing data integration density of our current computers. How will we handle the trillion sensor expectations?
For this scientists are working towards memristor technology as a viable option to replace current transistors. Memristors are not only able to store data but can process data as well and due to their ability to store and process in the same unit, the cost of energy is much less. Ferroelectric tunnel junctions are good memristors having very low current and fast operation speed (nanosecond time scale), making them extremely energy efficient. Our solution is a greener alternative to the complex oxide based devices proposed in recent times that require high temperature processing, contains hazardous metals like Lead, Barium, Bismuth etc. that can pose serious threat to human health when electronic components are discarded and dumped as land-fill wastes. Also our devices are extremely robust, can handle long and repeated operational stress and very energy efficient. So they have high potential for implementation a large-scale network.
You can also learn about Neuromorphic Computing from this Youtube video. This explains it nicely: