On Bootstrapping Machine Learning Performance Predictors via Analytical Models
This work addresses performance modeling for systems like key-value stores, but it appears incremental as it combines existing methodologies without claiming major breakthroughs.
The paper tackles the problem of performance modeling by reconciling white box and black box methodologies through a technique called Bootstrapping, and evaluates it via case studies on a Key-Value Store and a Total Order Broadcast service, though no concrete performance numbers are provided.
Performance modeling typically relies on two antithetic methodologies: white box models, which exploit knowledge on system's internals and capture its dynamics using analytical approaches, and black box techniques, which infer relations among the input and output variables of a system based on the evidences gathered during an initial training phase. In this paper we investigate a technique, which we name Bootstrapping, which aims at reconciling these two methodologies and at compensating the cons of the one with the pros of the other. We thoroughly analyze the design space of this gray box modeling technique, and identify a number of algorithmic and parametric trade-offs which we evaluate via two realistic case studies, a Key-Value Store and a Total Order Broadcast service.