Deep Reinforcement Learning for Join Order Enumeration

Marcus, Ryan; Papaemmanouil, Olga

doi:10.1145/3211954.3211957

Cited by 136 publications

(124 citation statements)

References 12 publications

Supporting

Mentioning

122

Contrasting

Unclassified

Order By: Relevance

“…As cardinality estimates improve, the need to re-optimize decreases. In Figure 8, we compare perfect-(n) plus re-optimization for varying values of n. We see that re-optimization improves the latency of perfect-(n) estimates until perfect- (5). While re-optimizing perfect-(5) slows the execution of the workload, the risk is relatively small, The execution time of the benchmark is only 6% slower with reoptimization than only perfect- (5).…”

Section: B Re-optimization and Better Cardinality Estimatesmentioning

confidence: 97%

“…Put simply, a small number of optimization mistakes leads to workload performance far below what is possible. Ideas to improve the optimizer in these cases include better selectivity estimates [2]- [4] providing more exotic search mechanisms for plans [5], [6], and changing query plans that have gone off the rails at runtime [7]- [11]. A more recent trend is to recast query optimization as a machine learning problem.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

How I Learned to Stop Worrying and Love Re-optimization

Perron¹,

Shang²,

Kraska³

2019

2019 IEEE 35th International Conference on Data Engineering (ICDE)

View full text Add to dashboard Cite

Cost-based query optimizers remain one of the most important components of database management systems for analytic workloads. Though modern optimizers select plans close to optimal performance in the common case, a small number of queries are an order of magnitude slower than they could be. In this paper we investigate why this is still the case, despite decades of improvements to cost models, plan enumeration, and cardinality estimation. We demonstrate why we believe that a re-optimization mechanism is likely the most cost-effective way to improve end-to-end query performance. We find that even a simple re-optimization scheme can improve the latency of many poorly performing queries. We demonstrate that re-optimization improves the end-to-end latency of the top 20 longest running queries in the Join Order Benchmark by 27%, realizing most of the benefit of perfect cardinality estimation.

show abstract

Section: B Re-optimization and Better Cardinality Estimatesmentioning

confidence: 97%

Section: Introductionmentioning

confidence: 99%

How I Learned to Stop Worrying and Love Re-optimization

Perron¹,

Shang²,

Kraska³

2019

2019 IEEE 35th International Conference on Data Engineering (ICDE)

View full text Add to dashboard Cite

show abstract

“…Deep learning We are not the first to apply deep learning to database management problems. Deep learning has seen a recent groundswell of activity in the systems community [55], including several works on query optimization [30,37], entity matching [35], index selection [40,45], indexes themselves [18], and cardinality estimation [17,27].…”

Section: Related Workmentioning

confidence: 99%

Plan-structured deep neural network models for query performance prediction

Marcus

Papaemmanouil

2019

Proc. VLDB Endow.

Self Cite

View full text Add to dashboard Cite

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.

show abstract

“…Together with two strategies of exploration and exploitation, the action a t will be selected. After that action, the agent will observe a new state S t+1 , and get the rewards R t of the new environment [18]. R t will be used to update the agent's strategies.…”

Section: A Deep Reinforcement Learningmentioning

confidence: 99%

Which Channel to Ask My Question?: Personalized Customer Service Request Stream Routing Using Deep Reinforcement Learning

Liu

Long

et al. 2019

IEEE Access

View full text Add to dashboard Cite

Customer services are critical to all companies, as they may directly connect to the brand reputation. Due to a great number of customers, e-commerce companies often employ multiple communication channels to answer customers' questions, for example, chatbot and hotline. On one hand, each channel has limited capacity to respond to customers' requests, on the other hand, customers have different preferences over these channels. The current production systems are mainly built based on business rules, which merely considers tradeoffs between resources and customers' satisfaction. To achieve the optimal tradeoff between resources and customers' satisfaction, we propose a new framework based on deep reinforcement learning, which directly takes both resources and user model into account. In addition to the framework, we also propose a new deep-reinforcement-learning based routing method -double dueling deep Q-learning with prioritized experience replay (PER-DoDDQN). We evaluate our proposed framework and method using both synthetic and a real customer service log data from a large financial technology company. We show that our proposed deep-reinforcement-learning based framework is superior to the existing production system. Moreover, we also show our proposed PER-DoDDQN is better than all other deep Q-learning variants in practice, which provides a more optimal routing plan. These observations suggest that our proposed method can seek the trade-off where both channel resources and customers' satisfaction are optimal.INDEX TERMS Deep Reinforcement Learning, Personalized Customer Service, Time-series data processing

show abstract

Deep Reinforcement Learning for Join Order Enumeration

Cited by 136 publications

References 12 publications

How I Learned to Stop Worrying and Love Re-optimization

How I Learned to Stop Worrying and Love Re-optimization

Plan-structured deep neural network models for query performance prediction

Which Channel to Ask My Question?: Personalized Customer Service Request Stream Routing Using Deep Reinforcement Learning

Contact Info

Product

Resources

About