Dataflow Mirroring: Architectural Support for Highly Efficient Fine-Grained Spatial Multitasking on Systolic-Array NPUs

We present dataflow mirroring, architectural support for low-overhead fine-grained systolic array allocation which overcomes the limitations of prior coarse-grained spatial-multitasking Neural Processing Unit (NPU) architectures. The key idea of dataflow mirroring is to reverse the dataflows of co-l...

Full description

Bibliographic Details

Published in:	2021 58th ACM/IEEE Design Automation Conference (DAC) pp. 247 - 252
Main Authors:	Lee, Jounghoo, Choi, Jinwoo, Kim, Jaeyeon, Lee, Jinho, Kim, Youngsok
Format:	Conference Proceeding
Language:	English
Published:	IEEE 05.12.2021
Subjects:	Artificial neural networks Design automation Hardware Multitasking Prototypes Resource management Systolic arrays
Online Access:	Get full text
Tags:	Add Tag No Tags, Be the first to tag this record!