16 February 2024 in Proceedings of the ACM on Measurement and Analysis of Computing Systems
Datasets of labeled network traces are essential for a multitude of machine learning (ML) tasks in networking, yet their availability is hindered by privacy and maintenance concerns, such as data staleness. To overcome this limitation, synthetic network traces can often augment existing datasets. Unfortunately, current synthetic trace generation methods, which typically produce only aggregated flow statistics or a few selected packet attributes, do not always suffice, especially when model training relies on having features that are only available from packet traces. This shortfall manifests in both insufficient statistical resemblance to real traces and suboptimal performance on ML tasks when employed for data augmentation. In this paper, we apply diffusion models to generate high-resolution synthetic network traffic traces. We present NetDiffusion1, a tool that uses a finely-tuned, controlled variant of a Stable Diffusion model to generate synthetic network traffic that is high fidelity and conforms to protocol specifications. Our evaluation demonstrates that packet captures generated from NetDiffusion can achieve higher statistical similarity to real data and improved ML model performance than current state-of-the-art approaches (e.g., GAN-based approaches). Furthermore, our synthetic traces are compatible with common network analysis tools and support a myriad of network tasks, suggesting that NetDiffusion can serve a broader spectrum of network analysis and testing tasks, extending beyond ML-centric applications.
1 April 2019 in Proceedings on Privacy Enhancing Technologies
Virtually every Internet communication typically involves a Domain Name System (DNS) lookup for the destination server that the client wants to communicate with. Operators of DNS recursive resolvers—the machines that receive a client’s query for a domain name and resolve it to a corresponding IP address—can learn significant information about client activity. Past work, for example, indicates that DNS queries reveal information ranging from web browsing activity to the types of devices that a user has in their home. Recognizing the privacy vulnerabilities associated with DNS queries, various third parties have created alternate DNS services that obscure a user’s DNS queries from his or her Internet service provider. Yet, these systems merely transfer trust to a different third party. We argue that no single party ought to be able to associate DNS queries with a client IP address that issues those queries. To this end, we present Oblivious DNS (ODNS), which introduces an additional layer of obfuscation between clients and their queries. To do so, ODNS uses its own authoritative namespace; the authoritative servers for the ODNS namespace act as recursive resolvers for the DNS queries that they receive, but they never see the IP addresses for the clients that initiated these queries. We present an initial deployment of ODNS; our experiments show that ODNS introduces minimal performance overhead, both for individual queries and for web page loads. We design ODNS to be compatible with existing DNS protocols and infrastructure, and we are actively working on an open standard with the IETF.
28 September 2023 in Proceedings of the ACM on Networking
Operational networks commonly rely on machine learning models for many tasks, including detecting anomalies, inferring application performance, and forecasting demand. Yet, model accuracy can degrade due to concept drift, whereby the relationship between the features and the target to be predicted changes. Mitigating concept drift is an essential part of operationalizing machine learning models in general, but is of particular importance in networking's highly dynamic deployment environments. In this paper, we first characterize concept drift in a large cellular network for a major metropolitan area in the United States. We find that concept drift occurs across many important key performance indicators (KPIs), independently of the model, training set size, and time interval---thus necessitating practical approaches to detect, explain, and mitigate it. We then show that frequent model retraining with newly available data is not sufficient to mitigate concept drift, and can even degrade model accuracy further. Finally, we develop a new methodology for concept drift mitigation, Local Error Approximation of Features (LEAF). LEAF works by detecting drift; explaining the features and time intervals that contribute the most to drift; and mitigates it using forgetting and over-sampling. We evaluate LEAF against industry-standard mitigation approaches (notably, periodic retraining) with more than four years of cellular KPI data. Our initial tests with a major cellular provider in the US show that LEAF consistently outperforms periodic and triggered retraining on complex, real-world data while reducing costly retraining operations.