OLAP (online analytical processing) is a present answer to decision support problems. OLAP technologies offer tools and concepts to perform arbitrary analysis efficiently from any kind of data. OLAP is not a single technology, it covers several concepts, including database organization, data presentation, and query modeling. The OLAP technology has attracted only minor interest from research community until recent years. Now the research is intensive. However, a large amount of commercial OLAP products has been already developed to fulfill increasing analysis needs.During the 1980’s business and government worked with data in the megabytes and gigabytes ranges. Contemporary enterprises have to manipulate data in range of terabytes and pedabytes. The need for more sophisticated analysis and faster synthesis of better quality information has grown. Today ‘s markets are much more competitive and dynamic than those are in past.
Business enterprise prospers or fails according to the sophistication and speed of their information system and their ability to analyze and synthesize information. Most notably lacking has been the ability to consolidate, view ,and analyze data, which makes sense to one or more specific enterprise analysts at given point in time. OLAP can be used for this purpose.
1.2Description of OLAP:
OLAP is a data analysis which has the capacity multidimensional data analysis and provide ways for the in which it can be aggregated, summarized, consolidated, summed, viewed and analyzed.
OLAP covers a wide spectrum of applications in which no single architecture could work optimally for the full range; data volumes range from few megabyte to terabytes; functionally ranges from simple aggregation to complex financial models;
Access ranges from read only to multi-user concurrent read/write;
User populations range from single user to thousand of users.
2.1.Types of OLAP:
On a technical basis, there are a surprisingly large number of variables that differentiate the product. No two products are identical Even architecture.
OLAP is divided into three types. They are
1.ROLAP (relational olap)
2.MOLAP (multidimensional olap)
3.HOALP (hybrid olap)
2.1.1Functions of ROLAP:
ROLAP uses SQL to extract data from a relational database and create multidimensional view on the fly. By managing relational databases systems with ROLAP we can offer powerful, simple solutions for void variety of commercial and scientific application problems. Relational systems are used in every industry for applications like storage, update retrieval, of information for operational, transaction, and complex processing as well as decision support systems, including query reporting.
In ROLAP model, multidimensional database server is replaced by a relational DBMS.Queries and database management can then be performed with a standard tools and interfaces like SQL and ODBC. Database can contain both summarized and detailed data. However, some concepts must be used to utilize preconsolidated data. Relational databases also have efficient methods to store null values, despite data is summarized or not. They also provide good scalability via techniques such as parallel query processing. The problem in ROLAP approach is that the large number of table required for multidimensional analysis degrades performance due to table joins and index processing. Another disadvantage of ROLAP is that analytical processing is done by the client program. However, separate servers can be used to offload analytical processing from the client.
MOLAP creates and stores multidimensional databases before hand for faster access. Multidimensional system analysis manipulate more easily and intuitively with in less time.
Multidimensional OLAP (MOLAP) and relational OLAP (ROLAP) [White 1996]. In the MOLAP model, the multidimensional database server (MDBMS) does both data management and analytical processing. Conceptually, data in an MDBMS is stored as a multidimensional array where each cell in the array is formed by the intersection of all the dimensions. In this approach, only a few of the cells have a value. Thus, handling Sparsity of the data becomes essential in the MOLAP model. Data is preconsolidated into MOLAP database, so any kind of queries which require summarized data can be responded efficiently. The problem is lack of standardization. Several MOLAP models have been presented, but none of them has yet received a wide acceptance. Another problem is scalability. Multidimensional models are good for handling summarized data but when more levels are needed for drilling down purposes, a size of the database Increases rapidly.
2.1.3Function of HOLAP:
OLAP Combines ROLAP and MOLAP by creating multidimensional that relational databases.
2.2Comparision between types of OLAP:
The terms ROLAP, MOLAP, HOLAP are approximation indicators of how the product works. In ROLAP active data is stored in RDBMS for analysis. In MOLAP data is stored in shared multidimensional database on serve in HOLAP data is stored in local data files on a client p.c. in ROLAP data consolidation and other calculations are done by multipass SQL in MOLAP data consolidation and other calculations are done by multi-user multidimensional application serve. In HOLAP they are done by the client p.c. itself.
2.3Differences between types of OLAP:
ROLAP tech cannot access the multidimensional data directly. The retrieval of data performance is very slow. It can handle large amount of data.
There are two concepts on which OLAP is based. They ar
1.Flexible information synthesis
2.Multiple data dimension / consolidation paths.
The synthesis of information from large databases is a task performed by all databases end-user and business analysts. The primary approach to synthesis of information from data is analysis.
“On-line analytical processing (OLAP)”, works almost exclusively with historical data deemed accurate as of a given point in time, specifically the brginning of the OLAP transaction.OLAP is made up of numerous, speculative “what-if”, and/or “why”, data model scenarios executed within the context of some specific historical basis a perspective. Within these scenarios, the values of key variables or parameters are changed, often repeatedly, to reflect potential variances in supply, production the economy, sells, market place, costs, and /or other environmental And internal factors. Information is than synthesized through animation of the data model.
Results of online analytic processing are normally displayed on terminals, but may have to be recorded in some of the data organization that are supported. When these results are stored in databases, it is necessary to be sure that speculative data in is kept separate from and not confused with data that represent the actual state of enterprise.
Data consolidation is the process of synthesizing pieces of information into single blocks of essential knowledge. The highest level in a data consolidation path is refereed to as that data dimension. There are typically a number of different.Dimensions from which a given pool of data can be analyzed.This plural perspective, or multi-dimensional conceptual view appears to be the way most business person naturally view there enterprise. Each of these perspectives is considered to be a complementary data dimension. Simultaneous analysis of multiple data dimensions is referred to as multi- dimension data analysis.Data consolidation paths consist series of consolidation Levels or steps that are defined in terms of multi-level parameters.these parameters apply to values from any variable where each successive level represents a higher degree of data consolidation. Thus for example, “business enterprise” could serve as a consolidation path for variable “worker”. “Business enterprise” Might include as a consolidation path the level “business area”, “division”,“department”,”project”,”task”, and “employee”. Data from the “employee” level could then be aggregated from the “task”, level to each more aggregated business organization level. If “task”, were sum of a variety of especially skilled subworkers such as plumbers, electrician, and lathe operator, then “skill” would serve as variable consolidation path.Consolidation may involve simple roll-ups or more complex relationship or equations and computations that span multiple consolidation paths or dimension. OLAP is name given to the dynamic enterprise analysis required to create, manipulate, animate, and synthesize information from exegetical, contemplative, and formulaic data analysis models. This includes the ability to discern new or anticipated relationships between variables.the ability to identify the parameters necessary to handle large amounts of data to create an unlimited number of dimensions (consolidation paths), and to specify cross-dimensional conditions and expressions.
No comments:
Post a Comment