HYDRA: Hybrid Data Multiplexing and Run-time Layer Configurable DNN Accelerator
This work addresses the problem of high hardware resource demands for DNNs on resource-constrained edge devices, representing an incremental improvement with specific optimizations.
The paper tackles the challenge of executing deep neural networks efficiently on edge devices with limited hardware resources by proposing HYDRA, a hybrid data multiplexing and runtime layer configurable DNN accelerator. The result includes reductions of over 90% in power consumption, resource utilization improvements, and 35.21 TOPSW performance.
Deep neural networks (DNNs) offer plenty of challenges in executing efficient computation at edge nodes, primarily due to the huge hardware resource demands. The article proposes HYDRA, hybrid data multiplexing, and runtime layer configurable DNN accelerators to overcome the drawbacks. The work proposes a layer-multiplexed approach, which further reuses a single activation function within the execution of a single layer with improved Fused-Multiply-Accumulate (FMA). The proposed approach works in iterative mode to reuse the same hardware and execute different layers in a configurable fashion. The proposed architectures achieve reductions over 90% of power consumption and resource utilization improvements of state-of-the-art works, with 35.21 TOPSW. The proposed architecture reduces the area overhead (N-1) times required in bandwidth, AF and layer architecture. This work shows HYDRA architecture supports optimal DNN computations while improving performance on resource-constrained edge devices.