Efficient Data-Driven Network Functions

TL;DR

This paper introduces Aquarius, a passive data collection framework that enhances network state inference in cloud environments using machine learning, achieving high accuracy and performance with minimal overhead.

Contribution

Aquarius enables efficient, low-overhead passive data collection for machine learning-based network state inference in virtual network functions.

Findings

Aquarius improves network state visibility significantly.

It demonstrates performance gains in traffic classification, autoscaling, and load balancing.

Low overhead is maintained during data collection and inference.

Abstract

Cloud environments require dynamic and adaptive networking policies. It is preferred to use heuristics over advanced learning algorithms in Virtual Network Functions (VNFs) in production becuase of high-performance constraints. This paper proposes Aquarius to passively yet efficiently gather observations and enable the use of machine learning to collect, infer, and supply accurate networking state information-without incurring additional signalling and management overhead. This paper illustrates the use of Aquarius with a traffic classifier, an autoscaling system, and a load balancer-and demonstrates the use of three different machine learning paradigms-unsupervised, supervised, and reinforcement learning, within Aquarius, for inferring network state. Testbed evaluations show that Aquarius increases network state visibility and brings notable performance gains with low overhead.