Answers are grounded in this paper and cite their sources.
High-performance edge artificial intelligence (Edge-AI) inference applications aim for high energy efficiency, memory density, and small form factor, requiring a design-space exploration across the whole stack—workloads, architecture, mapping, and co-optimization with emerging technology. In this article, we present a system-technology co-optimization (STCO) framework that interfaces with workload-driven system scaling challenges and physical design-enabled technology offerings. The framework is built on three engines that provide the physical design characterization, dataflow mapping optimizer, and system efficiency predictor. The framework builds on a systolic array accelerator to provide the design-technology characterization points using advanced imec A10 nanosheet CMOS node along with emerging, high-density voltage-gated spin-orbit torque (VGSOT) magnetic memories (MRAM), combined with memory-on-logic fine-pitch 3-D wafer-to-wafer hybrid bonding. We observe that the 3-D system integration of static random-access memory (SRAM)-based design leads to 9% power savings with 53% footprint reduction at iso-frequency with respect to 2-D implementation for the same memory capacity. Three-dimensional nonvolatile memory (NVM)-VGSOT allows <tex-math notation="LaTeX">4× </tex-math> memory capacity increase with 30% footprint reduction at iso-power compared with 2-D SRAM <tex-math notation="LaTeX">1× </tex-math>. Our exploration with two diverse workloads—image resolution enhancement (FSRCNN) and eye tracking (EDSNet)—shows that more resources allow better workload mapping possibilities, which are able to compensate peak system energy efficiency degradation on high memory capacity cases. We show that a 25% peak efficiency reduction on a <tex-math notation="LaTeX">32× </tex-math> memory capacity can lead to a <tex-math notation="LaTeX">7.4× </tex-math> faster execution with <tex-math notation="LaTeX">5.7× </tex-math> higher effective TOPS/W than the <tex-math notation="LaTeX">1× </tex-math> memory capacity case on the same technology.
No takes yet. Share an insight, caveat, or question.
Rocha et al. (2024) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: