For decades, database research has focused on optimizing performance under fixed resources. As more and more database applications move to the public cloud, we argue that it is time to make cost a first-class citizen when solving database optimization problems. In this paper, we introduce the concept of cost intelligence and envision the architecture of a cloud data warehouse designed for that. We investigate two critical challenges to achieving cost intelligence in an analytical system: automatic resource deployment and cost-oriented auto-tuning. We describe our system architecture with an emphasis on the components that are missing in today's cloud data warehouses. Each of these new components represents unique research opportunities in this much-needed research area.

翻译：数十年来，数据库研究领域始终致力于在固定资源条件下优化性能。随着越来越多的数据库应用迁移至公共云环境，我们认为在解决数据库优化问题时，应将成本视为首要考量因素。本文提出成本智能这一概念，并构想了一种专为此设计的云数据仓库架构。我们探究了在分析系统中实现成本智能所面临的两项关键挑战：自动资源部署与面向成本的自动调优。本文着重描述了当前云数据仓库中缺失的组件，并据此构建了我们的系统架构。这些新组件中的每一个都代表着这一亟需研究领域中的独特研究机遇。