Search for: All records

While recent years have witnessed a steady trend of applying Deep Learning (DL) to networking systems, most of the underlying Deep Neural Networks (DNNs) suffer two major limitations. First, they fail to generalize to topologies unseen during training. This lack of generalizability hampers the ability of the DNNs to make good decisions every time the topology of the networking system changes. Second, existing DNNs commonly operate as "blackboxes" that are difficult to interpret by network operators, and hinder their deployment in practice. In this paper, we propose to rely on a recently developed family of graph-based DNNs to address the aforementioned limitations. More specifically, we focus on a network congestion prediction application and apply Graph Attention (GAT) models to make congestion predictions per link using the graph topology and time series of link loads as inputs. Evaluations on three real backbone networks demonstrate the benefits of our proposed approach in terms of prediction accuracy, generalizability, and interpretability. 

Intrusion detection through classifying incoming packets is a crucial functionality at the network edge, requiring accuracy, efficiency and scalability at the same time, introducing a great challenge. On the one hand, traditional table-based switch functions have limited capacity to identify complicated network attack behaviors. On the other hand, machine learning based methods providing high accuracy are widely used for packet classification, but they typically require packets to be forwarded to an extra host and therefore increase the network latency. To overcome these limitations, in this paper we propose an architecture with programmable data plane switches. We show that Binarized Neural Networks (BNNs) can be implemented as switch functions at the network edge classifying incoming packets at the line speed of the switches. To train BNNs in a scalable manner, we adopt a federated learning approach that keeps the communication overheads of training small even for scenarios involving many edge network domains. We next develop a prototype using the P4 language and perform evaluations. The results demonstrate that a multi-fold improvement in latency and communication overheads can be achieved compared to state-of the-art learning architectures. 

WiFi is increasingly used by carriers for opportunistically offloading the cellular network infrastructure or even for increasing their revenue through WiFi-only plans and WiFi ondemand passes. Despite the importance and momentum of this technology, the current deployment of WiFi access points (APs) by the carriers follows mostly a heuristic approach. In addition, the prevalent free-of-charge WiFi access policy may result in significant opportunity costs for the carriers as this traffic could yield non-negligible revenue. In this paper, we study the problem of optimizing the deployment of WiFi APs and pricing the WiFi data usage with the goal of maximizing carrier profit. Addressing this problem is a prerequisite for the efficient integration of WiFi to next-generation carrier networks. Our framework considers various demand models that predict how traffic will change in response to alteration in price and AP locations. We present both optimal and approximate solutions and reveal how key parameters shape the carrier profit. Evaluations on a dataset of WiFi access patterns indicate that WiFi can indeed help carriers reduce their costs while charging users about 50% lower than the cellular service.