SLA-Centric Automated Algorithm Selection Framework for Cloud Environments

TL;DR

This paper introduces an SLA-aware automated algorithm selection framework for cloud environments, leveraging machine learning to optimize resource allocation and prevent SLA violations in combinatorial problems.

Contribution

It presents a novel ensemble machine learning framework that predicts and ranks algorithms based on SLA constraints in cloud settings, including a new dataset and empirical evaluation.

Findings

Effective prediction of algorithm performance under SLA constraints.

Improved algorithm selection accuracy using ensemble models.

Insights into hyperparameter and model impact on performance.

Abstract

Cloud computing offers on-demand resource access, regulated by Service-Level Agreements (SLAs) between consumers and Cloud Service Providers (CSPs). SLA violations can impact efficiency and CSP profitability. In this work, we propose an SLA-aware automated algorithm-selection framework for combinatorial optimization problems in resource-constrained cloud environments. The framework uses an ensemble of machine learning models to predict performance and rank algorithm-hardware pairs based on SLA constraints. We also apply our framework to the 0-1 knapsack problem. We curate a dataset comprising instance specific features along with memory usage, runtime, and optimality gap for 6 algorithms. As an empirical benchmark, we evaluate the framework on both classification and regression tasks. Our ablation study explores the impact of hyperparameters, learning approaches, and large language models effectiveness in regression, and SHAP-based interpretability.