一、OLTP、OLAP概念
OLTP
On-Line Transaction Processing联机事务处理过程(OLTP)
也称为面向交易的处理过程，其基本特征是前台接收的用户数据可以立即传送到计算中心进行处理，
并在很短的时间内给出处理结果，是对用户操作快速响应的方式之一。
On-Line Transaction Processing联机事务处理过程(OLTP)
也称为面向交易的处理过程，其基本特征是前台接收的用户数据可以立即传送到计算中心进行处理，
并在很短的时间内给出处理结果，是对用户操作快速响应的方式之一。
OLAP
On-line Analytical Processing 联机分析处理
它使分析人员能够迅速、一致、交互地从各个方面观察信息，以达到深入理解数据的目的。
On-line Analytical Processing 联机分析处理
它使分析人员能够迅速、一致、交互地从各个方面观察信息，以达到深入理解数据的目的。
二、OLTP、OLAP对比
联机分析处理 (OLAP) 的概念最早是由关系数据库之父E.F.Codd于1993年提出的，他同时提出了关于OLAP的12条准则。OLAP的提出引起了很大的反响，OLAP作为一类产品同联机事务处理 (OLTP) 明显区分开来。
当今的数据处理大致可以分成两大类：联机事务处理OLTP（on-line transaction processing）、联机分析处理OLAP（On-Line Analytical Processing）。OLTP是传统的关系型数据库的主要应用，主要是基本的、日常的事务处理，例如银行交易。
OLAP是数据仓库系统的主要应用，支持复杂的分析操作，侧重决策支持，并且提供直观易懂的查询结果。
下表列出了OLTP与OLAP之间的比较。
三、OLAP12准则
OLAP的12条准则 由关系数据库之父E.F.Codd于1993年提出的
1： Multidimensional conceptual view OLAP 模型必须提供多维概念视图
User-analysts would view an enterprise as being multidimensional in nature – for example, profits could be viewed by region, product, time period, or scenario (such as actual, budget, or forecast). Multi-dimensional data models enable more straightforward and intuitive manipulation of data by users, including "slicing and dicing".
分析用户能自然的视企业为一个多维模型，例如，利润可以按区域，产品，时间，或方案（如实际，预算或预测）查看。多维数据模型能让用户更直接和方便的操作数据，包括“切片和切块”
2： Transparency 透明性
When OLAP forms part of the users’ customary spreadsheet or graphics package, this should be transparent to the user. OLAP should be part of an open systems architecture which can be embedded in any place desired by the user without adversely affecting the functionality of the host tool.
The user should not be exposed to the source of the data supplied to the OLAP tool, which may be homogeneous or heterogeneous.
当OLAP以用户习惯的方式提供电子表格或图形显示时，这对用户应该是透明的。OLAP应该是开发系统架构的一部分，这个架构能按用户的需要嵌入到任何地方，而不会对主机工具的功能产生副作用。用户不应该接触到提供给OLAP工具的数据源，这些数据可能是同构的或是异构的
3： Accessibility 存取能力准则
The OLAP tool should be capable of applying its own logical structure to access heterogeneous sources of data and perform any conversions necessary to present a coherent view to the user. The tool (and not the user) should be concerned with where the physical data comes from.
OLAP工具应该有能力利用自有的逻辑结构访问异构数据源，并且进行必要的转换以提供给用户一个连贯的展示。是OLAP工具而不是用户需要关心物理数据的来源
4： Consistent reporting performance 稳定的报表能力
Performance of the OLAP tool should not suffer significantly as the number of dimensions is increased.
OLAP工具的性能不应该因维度增加而受到明显的影响
5： Client/server architecture 客户/服务器体系结构
The server component of OLAP tools should be sufficiently intelligent that the various clients can be attached with minimum effort. The server should be capable of mapping and consolidating data between disparate databases.
OLAP工具的服务器端应该足够的智能让多客户的以最小的代价连接。服务器应该有能力映射和巩固不同数据库的数据
6： Generic dimensionality 维的等同性准则
Every data dimension should be equivalent in its structure and operational capabilities.
每个数据维度应该具有等同的结构和操作能力
7： Dynamic sparse matrix handling 动态的稀疏矩阵处理
The OLAP server’s physical structure should have optimal sparse matrix handling.
OLAP服务器的物理结构应能处理最优稀疏矩阵
8： Multi-user support 多用户支持能力
OLAP tools must provide concurrent retrieval and update access, integrity and security.
OLAP应提供并发获取和更新访问，保证完整和安全的能力
9： Unrestricted cross-dimensional operations 非受限的跨维操作
Computational facilities must allow calculation and data manipulation across any number of data dimensions, and must not restrict any relationship between data cells.
计算设备必需允许跨数据维度的计算和数据操作，不能限制任何数据单元间的关系
10：Intuitive data manipulation 直观的数据操纵
Data manipulation inherent in the consolidation path, such as drilling down or zooming out, should be accomplished via direct action on the analytical model’s cells, and not require use of a menu or multiple trips across the user interface.
数据操作应在固定的路径下，例如钻或缩小，应该通过直接在分析模型的单元上完成，而不需要目录货多次的用户交互
11：Flexible reporting 灵活的报表生成
Reporting facilities should present information in any way the user wants to view it.
报表设备应该能以用户需要的任何方式展现信息
12：Unlimited dimensions and aggregation levels. 不受限的维与聚集层次
The number of data dimensions supported should, to all intents and purposes, be unlimited. Each generic dimensions should enable an essentially unlimited number of user-defined aggregation levels within any given consolidation path.
数据维度数量应该是无限的，用户在每个通用维度上定义的聚集聚合层次应该是无限的。
1： Multidimensional conceptual view OLAP 模型必须提供多维概念视图
User-analysts would view an enterprise as being multidimensional in nature – for example, profits could be viewed by region, product, time period, or scenario (such as actual, budget, or forecast). Multi-dimensional data models enable more straightforward and intuitive manipulation of data by users, including "slicing and dicing".
分析用户能自然的视企业为一个多维模型，例如，利润可以按区域，产品，时间，或方案（如实际，预算或预测）查看。多维数据模型能让用户更直接和方便的操作数据，包括“切片和切块”
2： Transparency 透明性
When OLAP forms part of the users’ customary spreadsheet or graphics package, this should be transparent to the user. OLAP should be part of an open systems architecture which can be embedded in any place desired by the user without adversely affecting the functionality of the host tool.
The user should not be exposed to the source of the data supplied to the OLAP tool, which may be homogeneous or heterogeneous.
当OLAP以用户习惯的方式提供电子表格或图形显示时，这对用户应该是透明的。OLAP应该是开发系统架构的一部分，这个架构能按用户的需要嵌入到任何地方，而不会对主机工具的功能产生副作用。用户不应该接触到提供给OLAP工具的数据源，这些数据可能是同构的或是异构的
3： Accessibility 存取能力准则
The OLAP tool should be capable of applying its own logical structure to access heterogeneous sources of data and perform any conversions necessary to present a coherent view to the user. The tool (and not the user) should be concerned with where the physical data comes from.
OLAP工具应该有能力利用自有的逻辑结构访问异构数据源，并且进行必要的转换以提供给用户一个连贯的展示。是OLAP工具而不是用户需要关心物理数据的来源
4： Consistent reporting performance 稳定的报表能力
Performance of the OLAP tool should not suffer significantly as the number of dimensions is increased.
OLAP工具的性能不应该因维度增加而受到明显的影响
5： Client/server architecture 客户/服务器体系结构
The server component of OLAP tools should be sufficiently intelligent that the various clients can be attached with minimum effort. The server should be capable of mapping and consolidating data between disparate databases.
OLAP工具的服务器端应该足够的智能让多客户的以最小的代价连接。服务器应该有能力映射和巩固不同数据库的数据
6： Generic dimensionality 维的等同性准则
Every data dimension should be equivalent in its structure and operational capabilities.
每个数据维度应该具有等同的结构和操作能力
7： Dynamic sparse matrix handling 动态的稀疏矩阵处理
The OLAP server’s physical structure should have optimal sparse matrix handling.
OLAP服务器的物理结构应能处理最优稀疏矩阵
8： Multi-user support 多用户支持能力
OLAP tools must provide concurrent retrieval and update access, integrity and security.
OLAP应提供并发获取和更新访问，保证完整和安全的能力
9： Unrestricted cross-dimensional operations 非受限的跨维操作
Computational facilities must allow calculation and data manipulation across any number of data dimensions, and must not restrict any relationship between data cells.
计算设备必需允许跨数据维度的计算和数据操作，不能限制任何数据单元间的关系
10：Intuitive data manipulation 直观的数据操纵
Data manipulation inherent in the consolidation path, such as drilling down or zooming out, should be accomplished via direct action on the analytical model’s cells, and not require use of a menu or multiple trips across the user interface.
数据操作应在固定的路径下，例如钻或缩小，应该通过直接在分析模型的单元上完成，而不需要目录货多次的用户交互
11：Flexible reporting 灵活的报表生成
Reporting facilities should present information in any way the user wants to view it.
报表设备应该能以用户需要的任何方式展现信息
12：Unlimited dimensions and aggregation levels. 不受限的维与聚集层次
The number of data dimensions supported should, to all intents and purposes, be unlimited. Each generic dimensions should enable an essentially unlimited number of user-defined aggregation levels within any given consolidation path.
数据维度数量应该是无限的，用户在每个通用维度上定义的聚集聚合层次应该是无限的。
整理自：
https://www.cnblogs.com/andy6/p/6011959.html
https://blog.csdn.net/lzhat/article/details/59102150