Processing-in-memory (PIM), as a novel computing paradigm, provides
significant performance benefits from the aspect of effective data movement
reduction. SRAM-based PIM has been demonstrated as one of the most promising
candidates due to its endurance and compatibility. However, the integration
density of SRAM-based PIM is much lower than other non-volatile memory-based
ones, due to its inherent 6T structure for storing a single bit. Within
comparable area constraints, SRAM-based PIM exhibits notably lower capacity.
Thus, aiming to unleash its capacity potential, we propose DDC-PIM, an
efficient algorithm/architecture co-design methodology that effectively doubles
the equivalent data capacity. At the algorithmic level, we propose a
filter-wise complementary correlation (FCC) algorithm to obtain a bitwise
complementary pair. At the architecture level, we exploit the intrinsic
cross-coupled structure of 6T SRAM to store the bitwise complementary pair in
their complementary states ($Q/\overline{Q}$), thereby maximizing the data
capacity of each SRAM cell. The dual-broadcast input structure and
reconfigurable unit support both depthwise and pointwise convolution, adhering
to the requirements of various neural networks. Evaluation results show that
DDC-PIM yields about $2.84\times$ speedup on MobileNetV2 and $2.69\times$ on
EfficientNet-B0 with negligible accuracy loss compared with PIM baseline
implementation. Compared with state-of-the-art SRAM-based PIM macros, DDC-PIM
achieves up to $8.41\times$ and $2.75\times$ improvement in weight density and
area efficiency, respectively.

提出了一种名为 DDC-PIM 的算法 / 架构协同设计方法，通过最大化每个 SRAM 单元的数据容量，有效地使等效数据容量翻倍，相较于 PIM 基准实现，DDC-PIM 在 MobileNetV2 上达到 2.84 倍加速，在 EfficientNet-B0 上达到 2.69 倍加速，无明显准确度损失，并且与最先进的基于 SRAM 的 PIM 宏相比，DDC-PIM 在重量密度和面积效率上分别达到 8.41 倍和 2.75 倍的改善。