arXiv subjects

Ryan Marcus

Publications and source records attributed to Ryan Marcus.

40 records · Page 3Linked to original sources

arXiv2019

Plan-Structured Deep Neural Network Models for Query Performance Prediction

Query performance prediction, the task of predicting the latency of a query, is one of the most challenging problem in database management systems. Existing approaches rely on features and performance models engineered by human experts, but often fail to capture the complex interactions between query operators and input relations, and generally do not adapt naturally to workload characteristics and patterns in query execution plans. In this paper, we argue that deep learning can be applied to the query performance prediction problem, and we introduce a novel neural network architecture for the task: a plan-structured neural network. Our approach eliminates the need for human-crafted feature selection and automatically discovers complex performance models both at the operator and query plan level. Our novel neural network architecture can match the structure of any optimizer-selected query execution plan and predict its latency with high accuracy. We also propose a number of optimizations that reduce training overhead without sacrificing effectiveness. We evaluated our techniques on various workloads and we demonstrate that our plan-structured neural network can outperform the state-of-the-art in query performance prediction.

cs.DB↗

arXiv2018

Towards a Hands-Free Query Optimizer through Deep Learning

Query optimization remains one of the most important and well-studied problems in database systems. However, traditional query optimizers are complex heuristically-driven systems, requiring large amounts of time to tune for a particular database and requiring even more time to develop and maintain in the first place. In this vision paper, we argue that a new type of query optimizer, based on deep reinforcement learning, can drastically improve on the state-of-the-art. We identify potential complications for future research that integrates deep learning with query optimization, and we describe three novel deep learning based approaches that can lead the way to end-to-end learning-based query optimizers.

cs.DB↗

arXiv2018

Deep Reinforcement Learning for Join Order Enumeration

Join order selection plays a significant role in query performance. However, modern query optimizers typically employ static join enumeration algorithms that do not receive any feedback about the quality of the resulting plan. Hence, optimizers often repeatedly choose the same bad plan, as they do not have a mechanism for "learning from their mistakes". In this paper, we argue that existing deep reinforcement learning techniques can be applied to address this challenge. These techniques, powered by artificial neural networks, can automatically improve decision making by incorporating feedback from their successes and failures. Towards this goal, we present ReJOIN, a proof-of-concept join enumerator, and present preliminary results indicating that ReJOIN can match or outperform the PostgreSQL optimizer in terms of plan quality and join enumeration efficiency.

cs.DB↗

arXiv2016

WiSeDB: A Learning-based Workload Management Advisor for Cloud Databases

Workload management for cloud databases must deal with the tasks of resource provisioning, query placement and query scheduling in a manner that meets the application's performance goals while minimizing the cost of using cloud resources. Existing solutions have approached these three challenges in isolation, and with only a particular type of performance goal in mind. In this paper, we introduce WiSeDB, a learning-based framework for generating holistic workload management solutions customized to application-defined performance metrics and workload characteristics. Our approach relies on supervised learning to train cost-effective decision tree models for guiding query placement, scheduling, and resource provisioning decisions. Applications can use these models for both batch and online scheduling of incoming workloads. A unique feature of our system is that it can adapt its offline model to stricter/looser performance goals with minimal re-training. This allows us to present alternative workload management strategies that address the typical performance vs. cost trade-off of cloud services. Experimental results show that our approach has very low training overhead while offering low cost strategies for a variety of performance goals and workload characteristics.

cs.DB↗