Learning to Configure Computer Networks with Neural Algorithmic Reasoning
This addresses the scalability challenge in network configuration for network administrators, offering a faster and more flexible approach compared to existing methods.
The paper tackles the problem of automatically configuring computer networks by relaxing the hard search problem into a learning-based objective, enabling configurations for much larger networks than before. The learned synthesizer is up to 490x faster than SMT-based methods and satisfies over 93% of requirements on average.
We present a new method for scaling automatic configuration of computer networks. The key idea is to relax the computationally hard search problem of finding a configuration that satisfies a given specification into an approximate objective amenable to learning-based techniques. Based on this idea, we train a neural algorithmic model which learns to generate configurations likely to (fully or partially) satisfy a given specification under existing routing protocols. By relaxing the rigid satisfaction guarantees, our approach (i) enables greater flexibility: it is protocol-agnostic, enables cross-protocol reasoning, and does not depend on hardcoded rules; and (ii) finds configurations for much larger computer networks than previously possible. Our learned synthesizer is up to 490x faster than state-of-the-art SMT-based methods, while producing configurations which on average satisfy more than 93% of the provided requirements.