Measurement driven modeling reveals unique generative AI network traffic patterns
MINT: Modeling GenAI Impact on Network Traffic
Networking and Internet Architecture
Summary
Generative AI, like chatbots and image generators, uses networks differently than older internet services. The paper presents MINT, a way to measure and simulate these new traffic patterns based on real data from multiple AI providers. It shows that AI traffic uploads and downloads are uneven and come in bursts that standard models don’t capture well. This new model helps networks prepare for and manage the growing AI-generated data flow more realistically.
What this means in practice
- •For network capacity planners: Use accurate AI traffic models to better predict and allocate network resources for emerging generative AI workloads.
- •For cloud infrastructure engineers: Simulate realistic generative AI traffic bursts to improve cloud server scheduling and load balancing strategies.
Authors
Andrew Nguyen, Samson Kempiak, Agrim Gupta, Koushik Kar, Ish Kumar Jain
Abstract
Generative AI (GenAI) is becoming a mainstream network workload, yet packet-level simulators lack measure\-ment-driven GenAI traffic models. Currently researchers must approximate GenAI services using traditional sources such as file transfer and video streaming, limiting realistic network evaluation of scheduling and capacity planning. We present MINT, a measurement and modeling framework for GenAI network traffic. Using an isolated net\-work-namespace capture pipeline, we collect client-side traces from three LLM providers across four modalities, cloud and edge servers, and wired and wireless network access points. We find that GenAI modalities exhibit distinct upload/download asymmetry and burst structures that differ from traditional applications. MINT clusters and models these burst regimes and validate empirical burst timing distribution behavior in ns-3 with normalized Wasserstein distances of 2--25\%. Our results also reveal that realistic packet bursts have significantly more variability than constant token generator models. MINT open-sources the first measurement-driven GenAI traffic model for packet-level network simulation.