The proposed AI processor architecture has high throughput for accelerating the neural network and reduces the external memory bandwidth required for processing the neural network. For achieving high throughput, the proposed super thread core (STC) includes 128 × 128 nano cores operating at the clock frequency of 1.2 GHz. The function‐safe architecture is proposed for a fault‐tolerance system such as an electronics system for autonomous cars. The general‐purpose processor (GPP) core is integrated with STC for controlling the STC and processing the AI algorithm. It has a self‐recovering cache and dynamic lockstep function. The function‐safe design has proved the fault performance has ASIL D of ISO26262 standard fault tolerance levels. Therefore, the entire AI processor is fabricated via the 28‐nm CMOS process as a prototype chip. Its peak computing performance is 40 TFLOPS at 1.2 GHz with the supply voltage of 1.1 V. The measured energy efficiency is 1.3 TOPS/W. A GPP for control with a function‐safe design can have ISO26262 ASIL‐D with the single‐point fault‐tolerance rate of 99.64%.

中文翻译：

---

具有ISO26262 ASIL‐D功能安全可编程多核的40‐TFLOPS人工智能处理器

所提出的AI处理器体系结构具有高吞吐量，可用于加速神经网络，并减少了处理神经网络所需的外部内存带宽。为了实现高吞吐量，建议的超线程内核（STC）包括以1.2 GHz时钟频率运行的128×128纳米内核。提出了一种功能安全体系结构，用于容错系统，例如自动驾驶汽车的电子系统。通用处理器（GPP）内核与STC集成在一起，用于控制STC和处理AI算法。它具有自动恢复缓存和动态锁步功能。功能安全的设计已证明其故障性能符合ISO26262标准容错等级的ASILD。因此，整个AI处理器是通过28 nm CMOS工艺制造的，作为原型芯片。在1.2 GHz的频率和1.1 V的电源电压下，其峰值计算性能为40 TFLOPS。测得的能量效率为1.3 TOPS / W。具有功能安全设计的用于控制的GPP可以具有ISO26262 ASIL‐D，其单点容错率为99.64％。