Generalized ping-pong staggers PIM macro write/compute operations to fully use off-chip bandwidth, outperforming naive ping-pong by 1.22x to 7.71x in simulation.
Metis AIPU: A 12nm 15TOPS/W 209.6TOPS SoC for cost-and energy-efficient inference at the edge,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.AR 1years
2024 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Generalized Ping-Pong: Off-Chip Memory Bandwidth Centric Pipelining Strategy for Processing-In-Memory Accelerators
Generalized ping-pong staggers PIM macro write/compute operations to fully use off-chip bandwidth, outperforming naive ping-pong by 1.22x to 7.71x in simulation.