Analysis Of Multidimensional Data Using Various Methods

  • Home
  • /
  • Technology
  • /
  • Analysis Of Multidimensional Data Using Various Methods

Analysis Of Multidimensional Data Using Various Methods

Nikhil Deshmukh
Abstract Data is exponentially increasing every year business wants to analyze data more accurately and efficiently Analyzing huge amounts of data is tedious tasks that involve considerable challenges commitments and organizational expense This paper provides an overview of different methods and tools to analyze data in the data warehouse We will analyze the six dimensional data using both relational database and multidimensional method and compare the performance by calculation using actual data

Keywords Data Warehouse Analysis OLAP Relational Multidimensional

Data warehouses contain data consolidated from several databases and are large in magnitude sometimes in terabytes Data warehouses are used mainly for decision ssupport applications and provide the summarized data than detailed individual records for analyzing purpose Some organizations are using data marts because data warehouse construction is a complicated process Data marts contain information in the form of subsets for any specific department On data warehouse and data marts different data analytical methods can be used In section II two methods of data analysis is explained first is conventional query method or using simple SQL and second method is Multidimensional analysis and its different types In the next section we have shown the incapability of conventional query method by taking the real world example and by comparing the performance of both on the basis of time taken to execute the operation and disk space used

A Query and Reporting

These are data query tools this type of tool formulate stand alone query and after analyzing statically it gives result in the form of graphs Such type of tools does not support multidimensional analysis and can execute only simple queries they do not offer aggregation and consolidation concepts These tools are optimum to generate request like How many number of articles do we have in

the stock 1 That is why these types of tools are called soft analysis tools

B Multidimensional analysis

In multidimensional database data is stored in the form of array table which allows fast visual representation of accumulated data Sometimes it is necessary to model data multidimensional for complex analysis and visualization especially in decision support system Multidimensional view or structure can be considered as cubes we can also call it cubes within cubes where each side of the cube is a dimension as shown in the fig A 4

To analyze multidimensional data OLAP Online analytical processing is used Types of OLAP are Multidimensional OLAP MOLAP Relational OLAP ROLAP Hybrid OLAP HOLAP and Spatial OLAP SOLAP

1 OLAP This type of server enables analysts to deep dive into performance through variety of view of the data It shows multidimensional phase of the business data through different views OLAP operations include Pivot change of orientation of the multidimensional view slice and dice selection and projection rollup increasing or decreasing the aggregation level along one or more dimensions Conceptual model of OLAP stress on aggregation as one of the key operation eg computing the total production by each state or by each month and give the ranking accordingly Some important characteristics of OLAP is summarization projected data fast interactive analysis multidimensional view Frequently changing business model and medium to large data sets 2



2 MOLAP Multidimensional OLAP directly supports the multidimensional view of the data through storage engine This provides very good indexing properties and speed but bad utilization of space especially in case of sparse data example is ESSBASE ARBOR 47

3 ROLAP Relational OLAP are the intermediate server sits between backend server and client It supports multidimensional OLAP query onthefly It utilizes transaction and scalability feature of relational system but mismatch between both queries can create performance issue 46

4 HOLAP Combination of MOL

* If you are the original writer of this essay and no longer wish to have the essay published on the WetPapers website then please email us at wet papers 1 @ gmail dot com for removal request.

© 2018 WetPapers. All rights reserved. Privacy Policy Terms of Service Disclaimer Copyright