Online Analytical Processing Types Computer Science Essay

Published: Last Edited:

This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.

OLAP is said to be online analytical processing System. This OLAP application is widely used by the organization attempts for business data which is available in large volumes of operational systems, spreadsheets, business partners. And it is decision support software used for analysing the business information and reports. It consists category of application and technology which allows a vast storage of data and also allows manipulation, reproduction of multidimensional data and manages the process. Business logic and statistical analysis for the user is said to be defined OLAP system. In this OLAP we can retrieve the data needed for the application user. The purpose of the OLAP is used to understand the OLAP concepts in that same concept we will have different names and different OLAP tools.

In OLAP relation data base as complex data and these data required two dimensional structures which does not support multidimensional views. OLAP software tool represents data in multidimensional formats which supports critical business questions in a company

These multidimensional views as technical business question for calculations and analyse require data from ware housing applications. The multidimensional data is store is also called as hyper cube. In this most relevantly used by the user is drill down and slice & dice

This OLAP are mainly divided into three types: Multidimensional OLAP (MOLAP) and Relational OLAP (ROLAP), Hybrid OLAP (HOLAP), Desktop Online analytical Processes (DOLAP)

MOLAP Multidimensional Online analytical Processes (Cubes)

ROLAP Relational Online analytical Processes (RDBMS)

HOLAP Combination of MOLAP & ROLAP

DOLAP Desktop Online analytical Processes (Cubes are stored in desktop also)

MOLAP (multidimensional online analytical process)

This is the most commonly used in OLAP Analysis. In MOLAP, multidimensional cube is which the data is said to be stored, but it is not in the relational database, in only proprietary formats. In this multidimensional database it as direct index of MOLAP(multidimensional online analytical process) is of online OLAP. In general the data is said to be treated as multidimensional in OLAP. They r different aspects or facets of data aggregates in sales by time and product model and these things are able view by the user, but in relation database the data which is stored can be viewed by multidimensional in only each dimensional of successively aspect of data aggregates. The data is said to be reflected when there is possible combinations of MOLAP data which is already stored in multidimensional array. The cube is accessed directly in each individual cell and for this MOLAP is mostly used for the faster response than the ROLAP (relational online analytical process) which is alternative to MOLAP

ROLAP (relational online analytical process)

This concept is completely based on generating the underline data which is stored in relational database to provide importance of OLAPs slicing and dicing functionality. The In SQL statement we have "WHERE" clause which is essence to each action of slicing and dicing functionality. In this relational database it has large amount of data which is supported by the ROLAP, but the main drawback is that it has performance slow in each report of SQL query in ROLAP but the functionality of SQL is tried to reduce the problem by implementing the tool out-of-the-box for complex problems and also providing the function for every user to define

HOLAP( hybrid online analytical process)

HOLAP is said to be hybrid online analytical process it is the mixture of both MOLAP and ROLAP. For summary-type information, HOLAP technology, for faster performance it uses leverages cube technology for which complete information is required, From cube the HOLAP can "drill through" the undergoing relational data. In this HOLAP the calculated data is generated before the cube is built, this is to apply easily the data querying but only for the less volume of data. In case of the large amount the data cannot retrieve from the cubes because data is pre- built, so it may be the deficiency in construction of cubes


The MEGASAVE is the large food retail chain and this MEGASAVE deals with large business data. In this MEGASAVE we have many business questions to be raised, so we should have large data. In all the OLAP's the MOLAP is more preferred most, because it deals with large data and we also need the large business data, so we need more processing speed for retrieving the data, MOLAP we have good processing speed by this we can retrieve the data efficiently. In this MOLAP productivity is more. These MOLAP have cubes these cubes are built for the fast retrieval of data, so the performance is high in this MOLAP. In this we deal with complex calculations. The cube is created with all pre-generated calculations. Hence complex calculation is not only achievable but can also return data quickly


This top down approach is nothing but the normalized database. In this it has cubes. The top down approach is a centralized database in one warehouse to different data mart. In this transfer of data is from different places of OLTP systems to centralized place to which the data is used to analyse. And the data is prearranged into subjected oriented, so that the data is easily accessible through atomic levels by drilling down. One data ware house consists of several data mart and these each of the data mart will built own department for the analyzing their data

In top down approach the data flow from the OLAP environment it starts with the data extraction from the data source and then that data is loaded into the production area and ensure a level of accuracy then transfer to ODS (operational data store). In operational data store sometimes it duplication in ODS. Data is overloaded into the data warehouse in a similar process to stay away from operational data store. Full data is extracted from temporary ODS which is storage aggregations and it extracted from data warehouse. These ODS is needed for the business




This bottom up approach is nothing but the De-normalized database. This approach will have star schema. Bottom up approach designed the data warehouse with the data mart which is connected in a bus structure. This approach is not a centralized and these will mainly linkage to the MEGASAVE. This data base can be used for the faster the processing speed. And it is different database. In this database we have fact tables and dimensional which uses queries and these queries also be fast processing in bottom up approach. To use the data marts we will have some common elements such, as dimensions and measure. From operational database data is extracted from the performance area. In that area data is loaded and recycle the data and store in the data marts. It has good balance between localized flexibility and centralised. Depending on the business requirements the ODS may not existing, so it increase complexity of process coordination


In these, two Top down approach and Bottom up approach. This BOTTOM UP approach will be specially designed for the OLAP browsing. And it is a de-normalized. We prefer bottom up approach for the MEGASAVE. In this bottom up approach we will use only star schema for all different business questions. In this bottom approach we can retrieve the data fast and also the details from the database. The importing thing is that we have very fast query processing with high speed. This is also very important and also efficient report presentation, because in big marts we have very large data, so these data should have a report. And this bottom up will be used for the business analysis in different ways. This is because it is mostly a dimensional database. In this dimensional data marts are reasonable stored in the database

But in case of The TOP DOWN approach it is all about the centralised database, so these will not use for the MEGASAVE for the large business data.


The relevance of architecture will change for different stores and also for the different MEGASAVE. For each store it has different data base in that case bottom is used because it is not centralised. In MEGASAVE we need multidimensional for this we use MOLAP because it supports the multidimensional. It enables the data ware manager to have a structure of what problem that are expected to emanate while designing the data ware. It gives a detail solution of what the enterprise MEGASAVE needs and also data ware house design is a complicated designing and hence it will give an architecture in order to reduce the problems to encountered the annotating the functions of some codes, fields before starting the real coding


Incomplete or missing values

Corrupted values

Out range of values

Wrong data

Duplicate data

Dissimilar data definitions

Data definitions varying overtime

Incompatible structures

Meaningless data

In data ware house we will find multiple types of data. And these data came from the different source, each source had a data and these data when they enter into the data ware house which is not appropriate. There will be also meaningless data this is nothing but a data which is not suitable and also not appropriate to the data ware house.

This is because, when the data enter into the data ware house from different source and sometimes it may not be correct data, these data contains some errors or problems. And problems might have different types, such as sometime data may be incomplete and also it may miss some values in data while transferring the data. It contains many problems like data is corrupted while entering into the data ware house it also has out range values which is not equal to the original value of the product

Some wrong data may enter into the source by mistake enter into the data it means that data will change into unreadable and the data will corrupted. In this data ware house we will also find duplication of data. This is will happen in the big data ware house, in data ware house it contains large data from different source so these may contain some duplicate data this is nothing but one data is said to be written in several times and it result to data duplications by this some data may be miss place because the those data and it will not appropriate

In data ware house we enter some data and at a certain point of time the data will changes the value at a time and this changes will not update in the data ware house and then problem will be raised. For example if we consider a product and the value of that product will be enter into the data ware house, at some point the value of product changes according into to the time varying over a period of time and that change of data may not be update into the data ware house, it will not so change the value of the product at that point of time the product shows previous data, so at that point of time the problem will raised. We can also find the problems like, when the data is said to be designed and if that data is transferred from one source to other source, sometimes the original data will not transfer in these case the problem may arise.


This WOLAP is nothing but a web based online analytical process. Web OLAP is a function of an OLAP tool from internet user, so this WOLAP tool supply OLAP functionality to drill-up and drill-down through dimensional hierarchy and good presentation and the benefits of the applications. This generally things about the problems which are faced by the web based OLAP. In this it first check the what ever the problems which are usually faced by the OLPA and this try to addresses the problems