BoGraph: Structured Bayesian Optimization From Logs for Systems with High-dimensional Parameter Space

Current auto-tuning frameworks struggle with tuning computer systems configurations due to their large parameter space, complex interdependencies, and high evaluation cost. Utilizing probabilistic models, Structured Bayesian Optimization (SBO) [18] has recently overcome these difficulties. SBO decomposes the parameter space by utilizing contextual information provided by system experts leading to fast convergence. However, the complexity of building probabilistic models has hindered its wider adoption. We propose BoGraph, a SBO framework that learns the system structure from its logs. BoGraph provides an API enabling experts to encode their knowledge of the system as performance models or components dependency. BoGraph takes in the learned structure and transforms it into a probabilistic graph model. Then it applies the expert-provided knowledge to the graph to further contextualize the system behavior. BoGraph probabilistic graph allows the optimizer to find efficient configurations faster than other methods.We evaluate BoGraph via a hardware architecture search problem, achieving an improvement in energy-latency objectives ranging from 5− 7 x-factors improvement over the default architecture. With its novel contextual structure learning pipeline, BoGraph makes using SBO accessible for a wide range of other computer systems such as databases and stream processors.