Computing In Spintronic Memory: A Thermal Perspective
This addresses thermal management for power-efficient CiM systems, which is an incremental but important domain-specific issue.
The paper tackles the thermal hotspot problem in Computing-in-Memory (CiM) systems by providing a quantitative thermal characterization, showing that temperature increases linearly with active memory cells and decreases with array size, with memory technology dictating thermal characteristics.
Computing-in-Memory (CiM) is a promising paradigm to address the memory bottleneck constraining traditional systems. Most power-efficient CiM variants can directly perform Boolean operations in non-volatile memory arrays. Higher microarchitectural activity due to CiM, however, can significantly increase power density (power per area) and result in thermal hotspots. In this paper, we provide a quantitative thermal characterization for CiM. We demonstrate that (i) the temperature remains mostly uniform due to lateral thermal conduction; (ii) the temperature increases linearly with the number of memory cells participating in computation; (iii) the temperature decreases linearly with the memory array size; (iv) the memory technology dictates the power density, hence the thermal characteristics.