Priority-Aware Model-Distributed Inference at Edge Networks
This work addresses efficient inference scheduling for edge networks with prioritized data sources, representing an incremental improvement over existing model-distributed methods.
The paper tackles the problem of model-distributed inference with multiple data sources having different priorities, formulating an optimization for model allocation and designing a PA-MDI algorithm that reduces inference time compared to baselines in experiments with edge devices and models like ResNet-50 and GPT-2.
Distributed inference techniques can be broadly classified into data-distributed and model-distributed schemes. In data-distributed inference (DDI), each worker carries the entire Machine Learning (ML) model but processes only a subset of the data. However, feeding the data to workers results in high communication costs, especially when the data is large. An emerging paradigm is model-distributed inference (MDI), where each worker carries only a subset of ML layers. In MDI, a source device that has data processes a few layers of ML model and sends the output to a neighboring device, i.e., offloads the rest of the layers. This process ends when all layers are processed in a distributed manner. In this paper, we investigate the design and development of MDI when multiple data sources co-exist. We consider that each data source has a different importance and, hence, a priority. We formulate and solve a priority-aware model allocation optimization problem. Based on the structure of the optimal solution, we design a practical Priority-Aware Model- Distributed Inference (PA-MDI) algorithm that determines model allocation and distribution over devices by taking into account the priorities of different sources. Experiments were conducted on a real-life testbed of NVIDIA Jetson Xavier and Nano edge devices as well as in the Colosseum testbed with ResNet-50, ResNet- 56, and GPT-2 models. The experimental results show that PA-MDI performs priority-aware model allocation successfully while reducing the inference time as compared to baselines.