Binary Neural Network (BNN) is a meaningful machine learning model on the data plane. However, due to the chip limitations, the scalability, especially the number of hidden layers in one pipeline, is limited. For better inference performance, existing methods reuse the hidden layers through packet recirculations. Recirculations lead to poor processing latency. Additionally, the simplified operations in the in-network BNN model restrict the flexibility of itself, which results in the unarbitrary input length of neurons for more the additional resource consumption than normal BNN model. In this paper, we present TBNN, an optimized in-network BNN model that achieves both scalability and flexibility. This approach eliminates deployment constraints while maximizing hardware utilization, advancing the feasibility of complex BNN models on resourcelimited data planes. By replacing computational bottleneck actions with Lookup Tables (LUTs), TBNN enables at most$4 \times$more neurons per pipeline and reduces per-packet latency by 50% through minimized recirculation. LUT-based implementation supports pruning operations, trading an accuracy loss of$\mathbf{1. 6 9 \%}$for saving about$\mathbf{2 4 \%}$instructions.

As the amount of data and complexity of neural network models continue to grow, distributed training has become increasingly crucial for improving training speed. However, the bottleneck of distributed training is the communication overheads among distributed workers. Recent research has shown that performing in-network aggregation using programmable switches is a good way to accelerate distributed training. However, previous work has only targeted specific neural network models and can only be applied in specified network topologies. Administrators may train different models and train them in different network topologies. In order to generalize the approach of using programmable switches to accelerate distributed training, we propose N4, a programmable intra-switch acceleration framework that supports distributed training of multiple neural networks. N4 also realizes the deployment of distributed workers based on any topology. Our experimental results show that N4 ensures high performance and isolation when training numerous neural networks. N4 outperforms state-of-the-art systems, accelerating training for existing methods by up to 3.4×.

The research of architecture has tremendous significance in realizing quantum Internet. Although there is not yet a standard quantum Internet architecture, the distributed architecture is one of the possible solutions, which utilizes quantum repeaters or dedicated entanglement sources in a flat structure for entanglement preparation & distribution. In this paper, we analyze the distributed architecture in detail and demonstrate that it has three limitations: 1) possible high maintenance overhead, 2) possible low-performance entanglement distribution, and 3) unable to support optimal entanglement routing. We design a hierarchical quantum Internet architecture and a communication model to solve the problems above. We also present a W-state Based Centralized Entanglement Preparation & Distribution (W-state Based CEPD) scheme and a Centralized Entanglement Routing (CER) algorithm within our hierarchical architecture and perform an experimental comparison with other entanglement preparation & distribution schemes and entanglement routing algorithms within the distributed architecture. The evaluation results show that the entanglement distribution efficiency of hierarchical architecture is 11.5% higher than that of distributed architecture on average (minimum 3.3%, maximum 37.3%), and the entanglement routing performance of hierarchical architecture is much better than that of a distributed architecture according to the fidelity and throughput.