Learning Queuing Networks by Recurrent Neural Networks
This provides a practical solution for practitioners in performance modeling who struggle with mathematical complexity, though it is incremental as it adapts existing neural network techniques to a specific domain.
The paper tackles the difficulty of building analytical performance models for queuing networks by proposing a machine-learning approach that encodes deterministic approximations into a recurrent neural network, resulting in interpretable white-box models with high predictive power demonstrated on synthetic and real load-balancing systems.
It is well known that building analytical performance models in practice is difficult because it requires a considerable degree of proficiency in the underlying mathematics. In this paper, we propose a machine-learning approach to derive performance models from data. We focus on queuing networks, and crucially exploit a deterministic approximation of their average dynamics in terms of a compact system of ordinary differential equations. We encode these equations into a recurrent neural network whose weights can be directly related to model parameters. This allows for an interpretable structure of the neural network, which can be trained from system measurements to yield a white-box parameterized model that can be used for prediction purposes such as what-if analyses and capacity planning. Using synthetic models as well as a real case study of a load-balancing system, we show the effectiveness of our technique in yielding models with high predictive power.