Algorithms For Efficient Top Down Join Enumeration
Download Algorithms For Efficient Top Down Join Enumeration PDF/ePub or read online books in Mobi eBooks. Click Download or Read Online button to get Algorithms For Efficient Top Down Join Enumeration book now. This website allows unlimited access to, at the time of writing, more than 1.5 million titles, including hundreds of thousands of titles in various foreign languages.
Algorithms for Efficient Top-Down Join Enumeration
Doctoral Thesis / Dissertation from the year 2014 in the subject Computer Science - Applied, grade: summa cum laude, University of Mannheim (School of Business Informatics and Mathematics), course: Databases, language: English, abstract: For a DBMS that provides support for a declarative query language like SQL, the query optimizer is a crucial piece of software. The declarative nature of a query allows it to be translated into many equivalent evaluation plans. The process of choosing a suitable plan from all alternatives is known as query optimization. The basis of this choice are a cost model and statistics over the data. Essential for the costs of a plan is the execution order of join operations in its operator tree, since the runtime of plans with different join orders can vary by several orders of magnitude. An exhaustive search for an optimal solution over all possible operator trees is computationally infeasible. To decrease complexity, the search space must be restricted. Therefore, a well-accepted heuristic is applied: All possible bushy join trees are considered, while cross products are excluded from the search. There are two efficient approaches to identify the best plan: bottom-up and top-down join enumeration. But only the top-down approach allows for branch-and-bound pruning, which can improve compile time by several orders of magnitude, while still preserving optimality. Hence, this thesis focuses on the top-down join enumeration. In the first part, we present two efficient graph-partitioning algorithms suitable for top-down join enumeration. However, as we will see, there are two severe limitations: The proposed algorithms can handle only (1) simple (binary) join predicates and (2) inner joins. Therefore, the second part adopts one of the proposed partitioning strategies to overcome those limitations. Furthermore, we propose a more generic partitioning framework that enables every graph-partitioning algorithm to handle join predicates involving more than two relations, and outer joins as well as other non-inner joins. As we will see, our framework is more efficient than the adopted graph-partitioning algorithm. The third part of this thesis discusses the two branch-and-bound pruning strategies that can be found in the literature. We present seven advancements to the combined strategy that improve pruning (1) in terms of effectiveness, (2) in terms of robustness and (3), most importantly, avoid the worst-case behavior otherwise observed. Different experiments evaluate the performance improvements of our proposed methods. We use the TPC-H, TPC-DS and SQLite test suite benchmarks to evaluate our joined contributions.
Efficient Memoization Algorithms for Query Optimization: Top-Down Join Enumeration through Memoization on the Basis of Hypergraphs
For a DBMS that provides support for a declarative query language like SQL, the query optimizer is a crucial piece of software. The declarative nature of a query allows it to be translated into many equivalent evaluation plans. The process of choosing a suitable plan from all alternatives is known as query optimization. The basis of this choice are a cost model and statistics over the data. Essential for the costs of a plan is the execution order of join operations in ist operator tree, since the runtime of plans with different join orders can vary by several orders of magnitude. An exhaustive search for an optimal solution over all possible operator trees is computationally infeasible. To decrease complexity, the search space must be restricted. Therefore, a well-accepted heuristic is applied: All possible bushy join trees are considered, while cross products are excluded from the search. There are two efficient approaches to identify the best plan: bottom-up and top- down join enumeration. But only the top-down approach allows for branch-and-bound pruning, which can improve compile time by several orders of magnitude, while still preserving optimality. Hence, this book focuses on the top-down join enumeration. In the first part, we present two efficient graph-partitioning algorithms suitable for top-down join enumer- ation. However, as we will see, there are two severe limitations: The proposed algo- rithms can handle only (1) simple (binary) join predicates and (2) inner joins. Therefore, the second part adopts one of the proposed partitioning strategies to overcome those limitations. Furthermore, we propose a more generic partitioning framework that enables every graph-partitioning algorithm to handle join predicates involving more than two relations, and outer joins as well as other non-inner joins. As we will see, our framework is more efficient than the adopted graph-partitioning algorithm. The third part of this book discusses the two branch-and-bound pruning strategies that can be found in the literature. We present seven advancements to the combined strategy that improve pruning (1) in terms of effectiveness, (2) in terms of robustness and (3), most importantly, avoid the worst-case behavior otherwise observed. Different experiments evaluate the performance improvements of our proposed methods. We use the TPC-H, TPC-DS and SQLite test suite benchmarks to evalu- ate our joined contributions. As we show, the average compile time [...]
Large-Scale Data Analytics
Author: Aris Gkoulalas-Divanis
language: en
Publisher: Springer Science & Business Media
Release Date: 2014-01-08
This edited book collects state-of-the-art research related to large-scale data analytics that has been accomplished over the last few years. This is among the first books devoted to this important area based on contributions from diverse scientific areas such as databases, data mining, supercomputing, hardware architecture, data visualization, statistics, and privacy. There is increasing need for new approaches and technologies that can analyze and synthesize very large amounts of data, in the order of petabytes, that are generated by massively distributed data sources. This requires new distributed architectures for data analysis. Additionally, the heterogeneity of such sources imposes significant challenges for the efficient analysis of the data under numerous constraints, including consistent data integration, data homogenization and scaling, privacy and security preservation. The authors also broaden reader understanding of emerging real-world applications in domains such as customer behavior modeling, graph mining, telecommunications, cyber-security, and social network analysis, all of which impose extra requirements for large-scale data analysis. Large-Scale Data Analytics is organized in 8 chapters, each providing a survey of an important direction of large-scale data analytics or individual results of the emerging research in the field. The book presents key recent research that will help shape the future of large-scale data analytics, leading the way to the design of new approaches and technologies that can analyze and synthesize very large amounts of heterogeneous data. Students, researchers, professionals and practitioners will find this book an authoritative and comprehensive resource.