You Only Charge Once 2.0 : A End-to-End Analog Computing-in-Memory Architecture with Reconfigurable Switched Capacitors
This work addresses the significant energy and area overhead of analog-to-digital converters in analog computing-in-memory accelerators, which is a critical problem for designers of energy-efficient AI hardware.
This paper introduces Charge-CIM, an analog computing-in-memory architecture that uses switched-capacitor charge redistribution for unified computing and conversion, thereby tackling the "ADC wall" bottleneck. The proposed architecture reduces ADC energy by 91.7% and improves energy efficiency by 2.7x and throughput by 2.0x compared to state-of-the-art charge-domain CIM accelerators.
Analog Computing-in-Memory (ACiM) accelerates deep neural networks by keeping weights inside memory arrays and executing dot products in the analog domain. However, modern ACiM accelerators are often limited by the "ADC wall": analog-to-digital converters consume a large fraction of energy and area, while bit-sliced execution repeatedly invokes these converters. Existing designs reduce this cost with low-resolution readout or time multiplexing, but they either lose output fidelity or introduce serialization overhead. Charge-CIM addresses this bottleneck by using switched-capacitor charge redistribution as a unified computing and conversion substrate. The same capacitor fabric performs input conversion, analog MAC, weighted shift-and-add, and readout quantization, reducing both standalone converter overhead and intermediate ADC invocations. A differential readout path further combines paired partial sums during ADC quantization, providing a highly compact and energy-efficient solution for array integration. With dataflow architecture support, we evaluated Charge-CIM on a suite of DNN benchmarks, from CNNs to Transformer models, and experimental results show that Charge-CIM reduces ADC energy by 91.7% under our evaluation setup and improves energy efficiency by 2.7x and throughput by 2.0x compared to the state-of-the-art charge-domain CIM accelerator.