In recent years, Spin-Transfer-Torque Magnetic Random Access Memory (STT-MRAM) has been considered as one of the most promising non-volatile memory candidates for in-memory computing. However, system-level performance gains using STT-MRAM for in-memory computing at deeply scaled nodes have not been assessed with respect to more mature memory technologies. In this letter, we present perpendicular magnetic tunnel junction (pMTJ) STT-MRAM devices at 28nm and 7nm. We evaluate the system-level performance of convolutional neural network (CNN) inference with STT-MRAM arrays in comparison to Static Random Access Memory (SRAM). We benchmark STT-MRAM and SRAM in terms of area, leakage power, energy, and latency from 65nm to 7nm technology nodes. Our results show that STT-MRAM keeps providing $\sim 5\times $ smaller synaptic core area, $\sim 20\times $ less leakage power, and $\sim 7\times $ less energy than SRAM when both devices are scaled from 65nm to 7nm. With the emerging need for low power computation for a broad range of applications such as internet-of-things (IoT) and neural network (NN), STT-MRAM can offer energy-efficient and high-density in-memory computing.

中文翻译：

---

用于内存计算的深度缩放自旋转移扭矩磁性随机存取存储器的性能前景

近年来，自旋-转移-转矩磁性随机存取存储器（STT-MRAM）被认为是最有前途的非易失性存储器候选之一，用于内存计算。然而，尚未针对更成熟的内存技术评估使用 STT-MRAM 在深度扩展节点上进行内存计算的系统级性能提升。在这封信中，我们介绍了 28nm 和 7nm 的垂直磁隧道结 (pMTJ) STT-MRAM 器件。与静态随机存取存储器 (SRAM) 相比，我们评估了使用 STT-MRAM 阵列的卷积神经网络 (CNN) 推理的系统级性能。我们在 65 纳米到 7 纳米技术节点的面积、泄漏功率、能量和延迟方面对 STT-MRAM 和 SRAM 进行了基准测试。我们的结果表明 STT-MRAM 不断提供 $\sim 5\times $ 较小的突触核心区， $\sim 20\times $ 更少的泄漏功率，和 $\sim 7\times $ 当两个器件从 65nm 缩放到 7nm 时，其能量比 SRAM 低。随着物联网 (IoT) 和神经网络 (NN) 等广泛应用对低功耗计算的新需求，STT-MRAM 可以提供节能和高密度的内存计算。