An adaptive range-query optimization technique with distributed replicas

Replication is an approach often used to speed up the execution of queries submitted to a large dataset.A compile-time/run-time approach is presented for minimizing the response time of 2-dimensional range when a distributed replica of a dataset exists.The aim is to partition the query payload(and its range) into subsets and distribute those to the replica nodes in a way that minimizes a client's response time.However,since query size and distribution characteristics of data(data dense/sparse regions) in varying ranges are not known a priori,performing efficient load balancing and parallel processing over the unpredictable workload is difficult.A technique based on the creation and manipulation of dynamic spatial indexes for query payload estimation in distributed queries was proposed.The effectiveness of this technique was demonstrated on queries for analysis of archived earthquake-generated seismic data records.

Query Optimization Framework for Graph Database in Cloud Dew Environment

The query optimizer uses cost-based optimization to create an execution plan with the least cost,which also consumes the least amount of resources.The challenge of query optimization for relational database systems is a combinatorial optimization problem,which renders exhaustive search impossible as query sizes rise.Increases in CPU performance have surpassed main memory,and disk access speeds in recent decades,allowing data compression to be used—strategies for improving database performance systems.For performance enhancement,compression and query optimization are the two most factors.Compression reduces the volume of data,whereas query optimization minimizes execution time.Compressing the database reduces memory requirement,data takes less time to load into memory,fewer buffer missing occur,and the size of intermediate results is more diminutive.This paper performed query optimization on the graph database in a cloud dew environment by considering,which requires less time to execute a query.The factors compression and query optimization improve the performance of the databases.This research compares the performance of MySQL and Neo4j databases in terms of memory usage and execution time running on cloud dew servers.

A Distributed DBMS Based Dynamic Programming Method for Query Optimization

Dynamic programming(DP) is an effective query optimization approach to select an appropriate join order for relational database management system(RDBMS) in multi-table joins. This method was extended and made available in distributed DBMS(D-DBMS). The structure of this optimal solution was firstly characterized according to the distributing status of tables and data, and then the recurrence relations between a problem and its sub-problems were recursively defined. DP in D-DBMS has the same time-complexity with that in centralized DBMS, while it has the capability to solve a much more sophisticated optimal problem of multi-table join in D-DBMS. The effectiveness of this optimal strategy has been proved by experiments.

An Efficient Algorithm for Query Transformation in Semantic Query Optimization

Semantic query optimization (SQO) is comparatively a recent approach for the transformation of given query into equivalent alternative query using matching rules in order to select an optimal query based on the costs of executing alternative queries. The key aspect of the algorithm proposed here is that previous proposed SQO techniques can be considered equally in the uniform cost model, with which optimization opportunities will not be missed. At the same time, the authors used the implication closure to guarantee that any matched rule will not be lost. The authors implemented their algorithm for the optimization of decomposed sub-query in local database in Multi-Database Integrator (MDBI), which is a multidatabase project. The experimental results verify that this algorithm is effective in the process of SQO.

Multi - weighted tree based query optimization method for parallel relational database systems

The author investigates the query optimization problem for parallel relational databases. A multi - weighted tree based query optimization method is proposed. The method consists of a multi - weighted tree based parallel query plan model, a cost model for parallel qury plans and a query optimizer. The parallel query plan model is the first one to model all basic relational operations, all three types of parallelism of query execution, processor and memory allocation to operations, memory allocation to the buffers between operations in pipelines and data redistribution among processors. The cost model takes the waiting time of the operations in pipelining execution into consideration and is computable in a bottom - up fashion. The query optimizer addresses the query optimization problem in the context of Select - Project - Join queries that are widely used in commercial DBMSs. Several heuristics determining the processor allocation to operations are derived and used in the query optimizer. The query optimizer is aware of memory resources in order to generate good - quality plans. It includes the heuristics for determining the memory allocation to operations and buffers between operations in pipelines so that the memory resourse is fully exploit. In addition, multiple algorithms for implementing join operations are consided in the query optimizer. The query optimizer can make an optimal choice of join algorithm for each join operation in a query. The proposed query optimization method has been used in a prototype parallel database management system designed and implemented by the author.

A solution of spatial query processing and query optimization for spatial databases

Recently, attention has been focused on spatial query language which is used to query spatial databases. A design of spatial query language has been presented in this paper by extending the standard relational database query language SQL. It recognizes the significantly different requirements of spatial data handling and overcomes the inherent problems of the application of conventional database query languages. This design is based on an extended spatial data model, including the spatial data types and the spatial operators on them. The processing and optimization of spatial queries have also been discussed in this design. In the end, an implementation of this design is given in a spatial query subsystem.

Learned Distributed Query Optimizer:Architecture and Challenges

The query processing in distributed database management systems(DBMS)faces more challenges,such as more operators,and more factors in cost models and meta-data,than that in a single-node DMBS,in which query optimization is already an NP-hard problem.Learned query optimizers(mainly in the single-node DBMS)receive attention due to its capability to capture data distributions and flexible ways to avoid hard-craft rules in refinement and adaptation to new hardware.In this paper,we focus on extensions of learned query optimizers to distributed DBMSs.Specifically,we propose one possible but general architecture of the learned query optimizer in the distributed context and highlight differences from the learned optimizer in the single-node ones.In addition,we discuss the challenges and possible solutions.

Processing and optimization of UMQL-based multimedia queries

Through the mapping from UMQL （ unified multimedia query language） conditional expressions to UMQA （unified multimedia query algebra） query operations, a translation algorithm from a UMQL query to a UMQA query plan is put forward, which can generate an equivalent UMQA internal query plan for any UMQL query. Then, to improve the execution costs of UMQA query plans effectively, equivalent UMQA translation formulae and general optimization strategies are studied, and an optimization algorithm for UMQA internal query plans is presented. This algorithm uses equivalent UMQA translation formulae to optimize query plans, and makes the optimized query plans accord with the optimization strategies as much as possible. Finally, the logic implementation methods of UMQA plans, i.e., logic implementation methods of UMQA operators, are discussed to obtain useful target data from a muifirnedia database. All of these algorithms are implemented in a UMQL prototype system. Application results show that these query processing techniques are feasible and applicable.

Research on the User Information Optimized Retrieval Technology in Distributed Database

The traditional method first classifies the user information and combines the query method to retrieve the interest information, but neglects to calculate the weight of the user interest information, which leads to the low retrieval accuracy. A retrieval method based on the fuzzy proximity classification technology is proposed. Approximation between the fuzzy sets is used to represent the consistency between the user interest information features, and the consistency calculation formula and the skewness confidence matrix between the user interest information features are given. The fuzzy classification of the user interest information can obtain the most consistent confidence data and eliminate the redundant approximation interference data. The probabilistic model of the information word frequency and the user interest information length calculates the weight of the user interest information, and adjusts the weight formula constantly.

An Improved Distributed Query for Large-Scale RDF Data

The rigid structure of the traditional relational database leads to data redundancy,which seriously affects the efficiency of the data query and cannot effectively manage massive data.To solve this problem,we use distributed storage and parallel computing technology to query RDF data.In order to achieve efficient storage and retrieval of large-scale RDF data,we combine the respective advantage of the storage model of the relational database and the distributed query.To overcome the disadvantages of storing and querying RDF data,we design and implement a breadth-first path search algorithm based on the keyword query on a distributed platform.We conduct the LUBM query statements respectively with the selected data sets.In experiments,we compare query response time in different conditions to evaluate the feasibility and correctness of our approaches.The results show that the proposed scheme can reduce the storage cost and improve query efficiency.

Dynamic Query Optimization Approach for Semantic Database Grid

Fundamentally, semantic grid database is about bringing globally distributed databases together in order to coordinate resource sharing and problem solving in which information is given well-defined meaning, and DartGrid II is the implemented database gird system whose goal is to provide a semantic solution for integrating database resources on the Web. Although many algorithms have been proposed for optimizing query-processing in order to minimize costs and/or response time, associated with obtaining the answer to query in a distributed database system, database grid query optimization problem is fundamentally different from traditional distributed query optimization. These differences are shown to be the consequences of autonomy and heterogeneity of database nodes in database grid. Therefore, more challenges have arisen for query optimization in database grid than traditional distributed database. Following this observation, the design of a query optimizer in DartGrid II is presented, and a heuristic, dynamic and parallel query optimization approach to processing query in database grid is proposed. A set of semantic tools supporting relational database integration and semantic-based information browsing has also been implemented to realize the above vision.

Query Optimization for Deductive Databases

A systematic, efficient compilation method for query evaluation of DeductiveDatabases (DeDB) is proposed in this paper. In order to eliminate redundancyand to minimize the potentially relevant facts, which are two key issues to theefficiency of a DeDB, the compilation process is decomposed into two phases.The first is the pre-compilation phase, which is responsible for the minimiza-tion of the potentially relevant facts. The second, which we refer to as thegeneral compilation phase, is responsible for the elimination of redundancy.The rule/goal graph devised by J. D. Ullman is appropriately extended andused as a uniform formalism. Two general algorithms corresponding to the twophases respectively are described intuitively and formally

PORLES:A Parallel Object Relational Database System

We developed a parallel object relational DBMS named PORLES. It uses BSP model as its parallel computing model, and monoid calculus as its basis of data model. In this paper, we introduce its data model, parallel query optimization, transaction processing system and parallel access method in detail.

GPU数据库实现技术发展演进

爆炸式增长的数据对存储和处理数据提出了更高的需求,GPU数据库作为新硬件数据库的一个重要分支,在大容量和高性能处理方面有其独特的优势.GPU数据库作为高性能数据库的代表,在最近几年受到学术界和产业界的关注,一批具有代表性的研究成果和标志性的实际产品已经出现.GPU数据库的技术发展按照GPU加速型和GPU内存型两种技术路线展开.两种技术路线都有相应的原型系统或产品出现.虽然两种GPU数据库的发展路线在实现上有所不同,但GPU数据库最基本的功能部分和核心技术是相似的,都有查询编译、查询优化、查询执行以及存储管理等功能.当前主流的数据传输方案除了PCIe之外,NVLink、RDMA和CXL等传输方案也为不同处理器之间的数据传输提供了更多的可能性.大多数GPU数据库使用列存储模型来存储数据,少数GPU数据库(如PG-Strom)对两种存储模型都支持.在列存储模型上利用压缩技术能减少数据的存储空间和传输时延.在GPU数据库上进行的压缩和解压的时间应该在整个数据处理的过程中占比很少.在GPU数据库上建立和维护索引不应该有很大的系统开销.JIT编译时间短、编译效率高,是GPU数据库编译的主流.操作符对数据库查询性能的影响非常明显,连接操作、分组聚集和OLAP运算符是目前研究最多的三个类型.目前大多数的研究中,连接和分组聚集算子通常结合在一起研究.在连接算子执行的过程中还和表的连接顺序结合在一起进行考虑.OLAP算子是GPU数据库中的又一个被大量研究的算子,GPU数据库在OLAP算子和模型方面持续受到研究者的关注.GPU数据库有三种查询处理模型,即行处理、列处理和向量化处理.向量化处理和列处理在实际系统中应用较多.由于GPU加速型数据库技术的发展,CPU-GPU协同处理模型上的查询方案与查询引擎也有一定数量的研究成果出现.当前GPU数据库的查询优化研究主要有三部分:多表连接顺序、查询重写和代价模型.然而,GPU数据库的代价评估模型在目前还没有很好的解决方案,GPU数据库的查询优化在未来仍有很大的研究空间.事务在GPU数据库中没有得到很好的研究,尽管有单独的原型系统,但目前的研究还没有取得重大进展.本文总结了GPU数据库各种关键技术已有的研究成果,指出GPU数据库当前存在的问题和面临的挑战,对未来的研究方向进行了展望.

面向分布式数据库的算子并行优化策略

随着网络技术的不断发展,数据规模呈现爆发式增长,使得传统的单机数据库逐步被分布式数据库所取代。分布式数据库采用节点协同工作方式解决了大规模数据存储问题,但由于增加了节点间通信开销,查询效率却不如单机数据库。分布式架构下,存储节点的数据仅用作多备份的冗余,为系统故障时提供数据恢复,并未被利用起来改善查询效率。针对上述问题,提出了一种面向分布式数据库的算子并行优化策略,通过对关键物理算子进行拆分,将拆分后的子请求均匀分配到存储层多个节点,由多个节点并行处理,从而减少查询响应时间。上述策略已经在分布式数据库CBase上进行了应用,实验表明,提出的并行优化策略可显著缩短SQL请求查询时间,并提高系统资源利用率。

基于仿真数据库的深潜球壳应力场数字孪生方法

提出一种基于仿真数据库及数字孪生技术的深潜球壳应力场预报方法,通过建立不同尺度及载荷下耐压球壳的应力场分布仿真数据库,在潜水器上实现了通过单个耐压壳有限传感器布点虚拟传感监测其他关键部位的应力状态。基于数字孪生技术构建三级虚拟结构层,Level-1 DT层实现从有限元仿真模型到数字模型的空间映射及云图展示,球壳的极限承载力实验与数值结果对比误差小于9.4%。Level-2 DT层通过创建数据库实现数字模型的数据样本推演,仿真数据库中未获得尺寸及载荷条件下的球壳应力场分布通过局部拉格朗日插值方法获得,插值应力结果相对于仿真结果的相对误差为4.8%。Level-3 DT层开发了深潜球壳数字模型危险区域应力场分布的机器学习预报功能,通过粒子群算法优化后的BP神经网络保证预测结果与仿真结果的误差小于1%。该方法综合考虑材料性能、结构尺寸和环境载荷,可以为耐压壳结构的实时安全评估提供参考,实现对潜水器单个或多个球壳动应力场分布的动态感知、智能诊断和科学预测。

面向Select和Sort的数据库算子缓存的设计与实现

缓存是数据库中提高查询性能的一种常用技术.目前,现有数据库缓存主要有两个方向:查询结果缓存和存储层块缓存.查询结果缓存是利用数据库查询执行的最终结果或中间结果(如子查询),而存储层块缓存则缓存查询涉及的底层数据块.本文从另外一个角度“缓存中含有的计算量”来重新审视缓存在查询优化中的应用,并以此为基础进一步划分数据库缓存方式.在查询执行过程中,数据库查询被转换成一系列操作(例如选择、排序等)的集合,而算子对应操作.查询处理中算子输出的数据为中间结果,含有部分计算量,我们将这部分数据进行缓存并加以利用.我们将这种缓存部分计算量的缓存方式称为算子缓存,即缓存每个操作执行后的结果.由于不同查询之间可能会存在相同算子,对相近数据执行相同计算,因此利用算子缓存加速查询执行性能具有相当大的潜力.本文的新颖之处在于从缓存含有的计算量角度出发,提出并研究算子缓存如何在查询优化中应用.本文以Filter、Sort算子为例,针对缓存复用提出了一种基于语义树的匹配算法,用于快速匹配缓存中的结果集.同时,针对复用缓存可能劣化查询性能的情况,提出使用基于成本的代价优化器防止使用缓存劣化查询性能.最后,本文基于开源分析型数据库ClickHouse实现了Filter、Sort算子缓存的原型,并对提出的算子缓存方案进行了大量的实验测试.结果表明,相比块缓存、物化视图方式,本文提出的算子缓存方案在本地SSD部署下最大能够分别提升9倍以及1.5倍的查询响应速度,在云环境下部署能够分别提升30倍以及2倍的查询响应速度.

基于机器学习的基数估计技术综述

基数估计是数据库关系系统查询优化器的基础和核心.随着人工智能技术的发展,其在数据处理、提取数据之间的关系等方面显现出优越的性能.近年来,基于机器学习的基数估计技术取得了显著的进展,受到了学术界的广泛关注.首先总结了基于机器学习的技术估计技术的发展现状,其次给出了基数估计的相关概念及其特征编码技术.接着建立了基数估计技术的分类体系.在此基础上,进一步将基于机器学习的基数估计技术细分为查询驱动、数据驱动和混合模型这3类基数估计技术.然后重点分析了每一类技术的建模流程、典型技术和模型特点,并对其在SQL和NoSQL中的应用进行了分析和总结.最后讨论了基于机器学习的基数估计技术面临的挑战和未来的研究方向。

基于TreeLSTM的查询基数估计

针对传统的数据库管理系统无法很好地学习谓词之间的交互以及无法准确地估计复杂查询的基数问题,提出了一种树形结构的长短期记忆神经网络(Tree Long Short Term Memory, TreeLSTM)模型建模查询,并使用该模型对新的查询基数进行估计.所提出的模型考虑了查询语句中包含的合取和析取运算,根据谓词之间的操作符类型将子表达式构建为树形结构,根据组合子表达式向量来表示连续向量空间中的任意逻辑表达式.TreeLSTM模型通过捕捉查询谓词之间的顺序依赖关系从而提升基数估计的性能和准确度,将TreeLSTM与基于直方图方法、基于学习的MSCN和TreeRNN方法进行了比较.实验结果表明:TreeLSTM的估算误差比直方图、MSCN、TreeRNN方法的误差分别降低了60.41%,33.33%和11.57%,该方法显著提高了基数估计器的性能.