Predictive Performance Comparison Analysis of Relational & NoSQL Graph Databases

Predictive Performance Comparison Analysis of Relational & NoSQL Graph Databases
复制标题

关系型预测性能比较分析

DOI：
--
复制
发表时间：
2017
期刊：
影响因子：
--
通讯作者：
W. Shahzad
W. Shahzad
中科院分区：
--
文献类型：
--
作者：
Wisal Khan;E. Ahmed;W. Shahzad

文献摘要

被引文献

相似文献

在过去的三十年里，关系数据库 在许多不同性质的组织中使用， 如教育、健康、商业和许多其他应用。 传统数据库表现出巨大的性能， 旨在使用ACID（原子性，一致性， 隔离性、持久性）属性来管理数据完整性。 在当今时代，组织正在存储更多的数据，即视频， 除了用于决策的结构化数据之外，还有图像、博客等。 同样，社交媒体和科学应用正在产生 大量不同性质的半结构化数据。关系 数据库无法正确处理和管理如此大量的 数据高效。为了解决这个问题，另一种模式 引入NoSQL数据库来管理和处理大量 大量的非结构化数据。nosql数据库是 分为四类，每一类都是根据 具体问题的性质和需要。本文 我们将比较Oracle关系数据库和NoSQL图 数据库使用优化查询和物理数据库调优 技术.比较是两个折叠：在第一次迭代中，我们 比较各种查询，如简单查询，数据库 调整Oracle关系数据库（如子数据库）， 在我们所需的环境中执行这些查询。其次对于 通过比较，我们将对结果进行预测分析 从我们的实验中得到的。
From last three decades, the relational databases are being used in many organizations of various natures such as Education, Health, Business and in many other applications. Traditional databases show tremendous performance and are designed to handle structured data with ACID (Atomicity, Consistency, Isolation, Durability) property to manage data integrity. In the current era, organizations are storing more data i.e. videos, images, blogs, etc. besides structured data for decision making. Similarly, social media and scientific applications are generating large amount of semi-structured data of varied nature. Relational databases cannot process properly and manage such large amount of data efficiently. To overcome this problem, another paradigm NoSQL databases is introduced to manage and process massive amount of unstructured data efficiently. NoSQL databases are divided into four categories and each category is used according to the nature and need of the specific problem. In this paper we will compare Oracle relational database and NoSQL graph database using optimized queries and physical database tuning techniques. The comparison is two folded: in the first iteration we compare various kinds of queries such as simpler query, database tuning of Oracle relational database such as sub databases and perform these queries in our desired environments. Secondly, for this comparison we will perform predictive analysis for the results obtained from our experiments.