Unicorn: Reasoning about Configurable System Performance through the lens of Causality

01/20/2022
by   Md Shahriar Iqbal, et al.
8

Modern computer systems are highly configurable, with the total variability space sometimes larger than the number of atoms in the universe. Understanding and reasoning about the performance behavior of highly configurable systems, over a vast and variable space, is challenging. State-of-the-art methods for performance modeling and analyses rely on predictive machine learning models, therefore, they become (i) unreliable in unseen environments (e.g., different hardware, workloads), and (ii) may produce incorrect explanations. To tackle this, we propose a new method, called Unicorn, which (i) captures intricate interactions between configuration options across the software-hardware stack and (ii) describes how such interactions can impact performance variations via causal inference. We evaluated Unicorn on six highly configurable systems, including three on-device machine learning systems, a video encoder, a database management system, and a data analytics pipeline. The experimental results indicate that Unicorn outperforms state-of-the-art performance debugging and optimization methods in finding effective repairs for performance faults and finding configurations with near-optimal performance. Further, unlike the existing methods, the learned causal performance models reliably predict performance for new environments.

READ FULL TEXT

page 6

page 18

research
02/26/2019

Transfer Learning for Performance Modeling of Configurable Systems: A Causal Analysis

Modern systems (e.g., deep neural networks, big data analytics, and comp...
research
06/13/2023

CAMEO: A Causal Transfer Learning Approach for Performance Optimization of Configurable Computer Systems

Modern computer systems are highly-configurable, with hundreds of config...
research
10/12/2020

CADET: A Systematic Method For Debugging Misconfigurations using Counterfactual Reasoning

Modern computing platforms are highly-configurable with thousands of int...
research
01/18/2023

CaRE: Finding Root Causes of Configuration Issues in Highly-Configurable Robots

Robotic systems have several subsystems that possess a huge combinatoria...
research
04/04/2019

Transfer Learning for Performance Modeling of Deep Neural Network Systems

Modern deep neural network (DNN) systems are highly configurable with la...
research
11/01/2019

Whence to Learn? Transferring Knowledge in Configurable Systems using BEETLE

As software systems grow in complexity and the space of possible configu...
research
06/25/2023

Utilizing deep learning for automated tuning of database management systems

Managing the configurations of a database system poses significant chall...

Please sign up or login with your details

Forgot password? Click here to reset