ILP Conference in Vienna 2019: Jennifer Rupp
April 23, 2019
Jennifer Rupp's presentation at the 2019 ILP Conference in Vienna.
Designing Memristor Materials and Functions for Neuromorphic Computing and Memories
Memristors are nano-devices that remember information permanently, switch in nanoseconds, are super dense, and power efficient. That makes memresistors potential replacements for good old transistors operated in DRAM, flash, and disk. What material architectures are used for memristor designs? How can we engineer their floor print and energy consumption? What if you can put huge amounts of storage near the processor and have enough bandwidth to exchange huge amounts of data? All at low power? Memristors are not just stuck in the past, they don't just remember, they can perform logic. And, the properties of the memristor apparently mimic neurons and can learn without supervision. The characteristics of memristors are such that you have to rethink the whole compute and storage paradigm.