To PiM or Not to PiM: The case for in-memory inferencing of quantized CNNs at the edge
Falcão, G.
; Ferreira, J.
Queue Vol. 20, Nº 6, pp. 9 - 34, December, 2022.
ISSN (print): 1542-7730
ISSN (online): 1542-7749
Scimago Journal Ranking: 0,42 (in 2022)
Digital Object Identifier: 10.1145/3580503
Download Full text PDF ( 3 MBs)
Downloaded 2 times
Abstract
As artificial intelligence becomes a pervasive tool for the billions of IoT (Internet of things) devices at the edge, the data movement bottleneck imposes severe limitations on the performance and autonomy of these systems. PiM (processing-in-memory) is emerging as a way of mitigating the data movement bottleneck while satisfying the stringent performance, energy efficiency, and accuracy requirements of edge imaging applications that rely on CNNs (convolutional neural networks).
