The exponential emergence of Field-Programmable Gate Arrays (FPGAs) has accelerated research on hardware implementation of Deep Neural Networks (DNNs). Among all DNN processors, domain-specific architectures such as Google’s Tensor Processor Unit (TPU) have outperformed conventional GPUs (Graphics Processing Units) and CPUs (Central Processing Units). However, implementing low-power TPUs in reconfigurable hardware remains a challenge in this field. Voltage scaling, a popular approach for energy savings, can be challenging in FPGAs, as it may lead to timing failures if not implemented appropriately. This work presents an ultra-low-power FPGA implementation of a TPU for edge applications. We divide the systolic array of a TPU into different FPGA partitions based on the minimum slack value of different design paths of Multiplier Accumulators (MACs). Each partition uses different near-threshold (NTC) biasing voltages to run its FPGA cores. The biasing voltage for each partition is roughly calculated by the proposed static schemes. However, further calibration of biasing voltage is performed by the proposed runtime scheme. To overcome the timing failure caused by NTC, the MACs with higher minimum slack are placed in lower-voltage partitions, while the MACs with lower minimum slack paths are placed in higher-voltage partitions. The proposed architecture is implemented in a commercial platform, namely with Xilinx FPGA and academic platform with 22 nm, 45 nm and 130 nm FPGAs. Any timing error caused by NTC can be caught by the Razor flipflop used in each MAC. The proposed voltage-scaled, partitioned systolic array can save 3.1% to 11.6% of dynamic power in and tools, respectively, depending on the FPGA technology, partition size, number of partitions and biasing voltages. The normalized performance and accuracy of benchmark models running on our low-power TPU are very competitive compared to existing literature.
The exponential emergence of Field Programmable Gate Array (FPGA) has accelerated the research of hardware implementation of Deep Neural Network (DNN). Among all DNN processors, domain specific architectures, such as, Google's Tensor Processor Unit (TPU) have outperformed conventional GPUs. However, implementation of TPUs in reconfigurable hardware should emphasize energy savings to serve the green computing requirement. Voltage scaling, a popular approach towards energy savings, can be a bit critical in FPGA as it may cause timing failure if not done in an appropriate way. In this work, we present an ultra low power FPGA implementation of a TPU for edge applications. We divide the systolic-array of a TPU into different FPGA partitions, where each partition uses different near threshold (NTC) biasing voltages to run its FPGA cores. The biasing voltage for each partition is roughly calculated by the proposed static schemes. However, further calibration of biasing voltage is done by the proposed runtime scheme. Four clustering algorithms based on the minimum slack value of different design paths of Multiply Accumulates (MACs) study the partitioning of FPGA. To overcome the timing failure caused by NTC, the MACs which have higher minimum slack are placed in lower voltage partitions and the MACs have lower minimum slack path are placed in higher voltage partitions. The proposed architecture is simulated in a commercial platform : Vivado with Xilinx Artix-7 FPGA and academic platform VTR with 22nm, 45nm, 130nm FPGAs. The simulation results substantiate the implementation of voltage scaled TPU in FPGAs and also justifies its power efficiency.