Cruise Control: Dynamic Model Selection for ML-Based Network Traffic Analysis

2024-12-19

arXiv.org

AI Summary

Existing machine learning model deployment methods for real-time network traffic analysis neglect dynamic network conditions and system resource constraints, leading to degraded accuracy and resource overloading. Method: We propose an online, system-driven dynamic model selection mechanism that enables millisecond-scale model switching based on lightweight system-state signals (e.g., CPU utilization, queueing latency), departing from conventional offline, static model selection paradigms. Our approach employs a pre-trained model library spanning multiple accuracy–cost trade-offs and integrates real-time load awareness with a retraining-free adaptive decision strategy. Contribution/Results: Evaluated on real-world traffic tasks, our method improves median accuracy by 2.78%, reduces packet loss rate to 25% of the baseline, and significantly enhances analytical robustness and operational availability under stringent resource constraints.

Technology Category

• Machine Learning: Learning on the Edge & Model Compression
• Natural Language Processing: Interpretability, Analysis, and Evaluation of NLP Models
• Planning, Routing, and Scheduling: Model-Based Reasoning

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems

Abstract

Modern networks increasingly rely on machine learning models for real-time insights, including traffic classification, application quality of experience inference, and intrusion detection. However, existing approaches prioritize prediction accuracy without considering deployment constraints or the dynamism of network traffic, leading to potentially suboptimal performance. Because of this, deploying ML models in real-world networks with tight performance constraints remains an open challenge. In contrast with existing work that aims to select an optimal candidate model for each task based on offline information, we propose an online, system-driven approach to dynamically select the best ML model for network traffic analysis. To this end, we present Cruise Control, a system that pre-trains several models for a given task with different accuracy-cost tradeoffs and selects the most appropriate model based on lightweight signals representing the system's current traffic processing ability. Experimental results using two real-world traffic analysis tasks demonstrate Cruise Control's effectiveness in adapting to changing network conditions. Our evaluation shows that Cruise Control improves median accuracy by 2.78% while reducing packet loss by a factor of four compared to offline-selected models.

Problem

• Dynamically selects best ML model for network traffic analysis
• Addresses deployment constraints and network traffic dynamism
• Optimizes accuracy-cost tradeoffs under performance constraints

Innovation

• Online dynamic model selection system
• Pre-trains multiple accuracy-cost tradeoff models
• Uses lightweight signals for traffic adaptation