OLAP technologies. Analytical systems OLAP Information technologies for data analysis olap

The minds of high competition and the dynamics of the current environment dictate advances in business management systems. The development of management theory and practice was accompanied by the emergence of new methods, technologies and models focused on increasing the efficiency of business. Methods and models have been adopted by the emergence of analytical systems. The demand for analytical systems in Russia is high. What appears to be a stagnant system in the financial sector: banks, insurance business, investment companies. The results of the work of analytical systems are necessary for us first of all for people, especially those who are responsible for the development of the company: specialists, experts, analysts. Analytical systems allow for the development of consolidation, visibility, optimization and forecasting. Until now, there has not been a residual classification of analytical systems, since there is no primary system for the meaning of the terms that are used in this direction. The information structure of an enterprise can be represented by a sequence of levels, each of which is characterized by its way of processing and managing information, which has its function in the management process. Thus, analytical systems will be deployed hierarchically at different levels of the infrastructure.

Range of transaction systems

Rhubarb of data

Showcase data

Riven OLAP – systems

Range of analytical programs

OLAP - systems - (OnLine Analytical Processing, analytical processing in the present day) - a technology for comprehensive data analysis. OLAP is a system primarily dedicated to the analysis of multi-factor data. An effective way to analyze and generate sounds. Looked at data warehouses, data windows and OLAP systems linked to Business Intelligence (BI) systems.

Most often, information and analytical systems, created for the immediate use of decision-makers, are found to be very simple in their design, but rather severely limited in functionality. Such static systems are called in the literature Kerivnik Information Systems (ISR) or Executive Information Systems (EIS). There will be a perception of the impersonality of the drinks and, being sufficient for everyday inspection, unreleased information on all meals until obvious data that may appear when a decision is made. The result of the operation of such a system is the production of rich signals, after the careful processing of which by the analyst, a new series of data appears. However, the new entry, failure to transfer during the design of such a system, is due primarily to formal descriptions, coding by the programmer and only then modifications. The hour of recuperation for such a time may become more and more days, but it will never again be pleasant. Thus, the extreme simplicity of static DSS, as most managers of information and analytical systems are actively fighting, turns into a catastrophic waste of power.

Dynamic DSS, however, is aimed at processing ad hoc queries from analysts to data. The greatest insight into such systems is by looking at E. F. Codd's statistics, which pioneered the concept of OLAP. The work of analysts with these systems involves an interactive sequence of forming queries and analyzing their results.

Alternatively, dynamic DSS can operate in the field of online analytical processing (OLAP); Support for making management decisions based on accumulated data can be focused on three basic areas.

Scope of detailed data. This area is influxing a large number of systems that are oriented towards searching for information. Most relational DBMSs cope well with any problems that arise here. The leading standard for manipulating relational data is SQL. Information-search systems that provide the interface of the end-user in tasks of searching for detailed information can be used both over other databases of these transactional systems and over the underground treasure of tributes.

The sphere of aggregated indicators. A comprehensive look at the information collected in the data warehouse, its organized and aggregation, hypercube display and rich analysis of the tasks of online analytical processing (OLAP) systems. Here you can either focus on special rich DBMSs or rely on relational technologies. In another type, aggregated data can be collected from a distance in a highly visible database, and aggregation of information can be carried out during the process of scanning the detailed table of a relational database.

The sphere of patterns. Intelligent processing is carried out by methods of intelligent data analysis (IDA, Data Mining), the main tasks of which are the search for functional and logical patterns in the accumulation of information, based on models and rules that explain the detection of anomalies and/or forecast the comfort of developments in current processes.

Operational analytical compilation of data

The OLAP concept is based on the principle of a rich and diverse representation of data. In 1993, the article by EF Codd looked at the fragments of the relational model, having previously pointed out the impossibility of “combining, reviewing and analyzing data from the point of view of multiplicity of changes, so that it is most meaningful for corporate analysis This way", and having identified hidden capabilities to OLAP systems, which expands functionality relational DBMS and which includes a rich analysis as one of its characteristics.

Classification of OLAP products based on the method of data presentation.

There are currently a large number of products on the market that provide OLAP functionality in both this and other worlds. Nearly 30 of the largest individuals were re-insuranced from the list of the surveillance Web server http://www.olapreport.com/. Providing a rich conceptual understanding from the backend interface to the output database, all OLAP products are divided into three classes based on the type of output database.

The largest operational analytical processing systems (for example, Essbase from Arbor Software, Oracle Express Server from Oracle) were up to the MOLAP class, so they could work with their own large databases. They rely on patented technologies for the world's largest and most expensive DBMSs. These systems will ensure the latest OLAP processing cycle. You can either include, in addition to the server component, the power of integration of the client interface, or use a vicor to link with external programs and spreadsheets. To maintain such systems, a special staff of service technicians is required, who are involved in the installation, maintenance of the system, and the formation of data on the terminal components.

Online Relational Analytical Processing (ROLAP) systems allow data stored in a relational database to be presented in a rich world form, ensuring the information is seamlessly processed into a rich world model through an interconnected meta tributes ROLAP systems are well suited to work with great advantages. Similar to MOLAP systems, they generate significant costs for the maintenance of information technology workers and transfer costs to a richly automated system of work.

We have decided that hybrid systems (Hybrid OLAP, HOLAP) will be decomposed using the method of increasing the advantage and minimizing the underprivileged power of the upper classes. What class belongs to the Media/MR company Speedware. According to the vendors, this will provide the analytical flexibility and flexibility of the MOLAP platform with instant access to real data powered by ROLAP.

Rich World OLAP (MOLAP)

In specialized DBMSs, based on a rich data supply, the data is organized in the form of a relational table, and looks like ordering a rich data array:

1) hypercubes (all are saved in the database in the middle of the guilty mother, however, in order to remain in the most complete database of extinctions) or

2) polycubes (skin changes are saved from the moisture set and all processing difficulties associated with this are transferred to the internal mechanisms of the system).

The use of rich databases in operational analytical processing systems has the same advantages.

In the case of a wide variety of DBMSs, the search and selection of data is much faster, less so with a rich conceptual view of the relational database, since the rich data base is denormalized, in advance aggregated displays and ensures optimized access to the middle that will be powered.

A wide variety of DBMSs can easily cope with the tasks of including a variety of functions to the information model, while objectively using SQL databases to solve these tasks based on relational DBMSs are complex and sometimes intractable.

On the other hand, there is a substantive exchange.

Database databases do not allow you to work with large databases. In addition, due to the denormalization and previously established aggregation of data from a large global database, it usually corresponds (according to Codd’s estimate) to 2.5-100 times less detailed data output.

Rich DBMSs compared to relational ones do not effectively use external memory. For the most important types of information, the hypercube is very sparse, and the remaining data is stored in an ordered form; unimportant values can only be determined by selecting the optimal sorting order. allows you to organize data in groups as seamlessly as possible. However, in this case, the problem occurs less often. In addition, the optimal sorting order from the point of view of saving rare data is completely inconsistent with the order that is most often used in queries. Therefore, in real systems there is a trade-off between the speed code and the large amount of disk space occupied by the database.

Well, many of the world's DBMSs have been rewarded even more for such minds.

The amount of output data for analysis is small (no more than a few gigabytes), so the rate of data aggregation is high.

The set of information changes is stable (any change in their structure may in the future require a new awakening of the hypercube).

The hour the system responds to unregulated power supply is a critical parameter.

There is a need for a wider range of folding functions to calculate large calculations over the midpoints of the hypercube, including the possibility of writing koristuvach functions.

Relational OLAP (ROLAP)

The immediate use of relational databases in operational analytical processing systems has the same benefits.

Most corporate data collections are implemented using relational DBMSs, and ROLAP tools allow analysis directly from them. In this case, the size of the tendon is not such a critical parameter as in the MOLAP type.

For different dimensions, if changes to the structure have to be made frequently, ROLAP systems with dynamic dimensions and optimal solutions, since such modifications do not affect physically ї database reorganization.

Relational DBMSs provide significantly higher levels of data protection and good ability to differentiate access rights.

The main shortcoming of ROLAP in a world with a rich DBMS is less productivity. To ensure the productivity that comes with MOLAP, relational systems require careful processing of database schemas and adjustment of indexes, which greatly contributes to the efforts of database administrators. Only with the use of highly efficient schemes, the productivity of well-tuned relational systems can be close to the productivity of systems based on the world's richest databases.

In 1993, the founder of the relational approach to databases, Edgar Codd and his partners (Edgar Codd, a mathematician and IBM Fellow), published an article initiated by Arbor Software (today a leading company I "Hyperion Solutions"), entitled "Both analytical reports) for business analysts", which formulated 12 features of OLAP technology, which later resulted in an additional six. These provisions have become the main basis of new and even promising technology.

Main features of the technology OLAP (Basic):

there is a rich variety of conceptual representations of these data;
intuitive manipulation of data;
availability and detail of data;
package not collection of tributes against interpretation;
OLAP analysis models;
"client-server" architecture (OLAP accessible from the desktop);
insight (insightful access to external data);
insured for a rich supply of support.

Special features(Special):

processing of informal data;
saving OLAP results: saving them separately as output data;
turning off daily values;
processing of daily values.

Features of the presentation of calls(Report):

flexibility of molding;
standard productivity of people;
automatic adjustment of the physical level of data acquisition.

Controlling the Worlds(Dimension):

the universality of extinctions;
the amount of vibration and similar aggregation cannot be determined;
The number of operations between dimensions is not limited.

Historically, it has happened that today's term "OLAP" is respected not only by the rich view of the data from the side of the end user, but also by the rich representation of data in the entire database. The appearance of independent terms is connected with this "Relational OLAP"(ROLAP) "Rich OLAP"(MOLAP).

OLAP service is a tool for analyzing large amounts of data in real time. Interacting with the OLAP system, you can perform a detailed review of information, extract additional data and perform analytical operations such as detailing, sampling, cross-section Ilu, the clock changes overnight due to a wide variety of parameters. All work with the OLAP system is carried out in terms of the subject area and allows for a statistically based model of the business situation.

OLAP software features - This is a tool for rapid data analysis, what to take place at the shelter. The main feature is that they are oriented towards the development of not a specialist in information technology, not an expert statistician, but a professional in applied management - a manager of a department, a department, She will be named director. Kosts are recognized for spilting analysis with the problem, not with the computer. In Fig. 6.14 readings is an elementary OLAP cube that allows you to evaluate data in three dimensions.

A rich OLAP cube and a system of advanced mathematical algorithms for statistical processing allow you to analyze data of any complexity at any time intervals.

Rice. 6.14.

Being aware of his well-ordered mechanisms for manipulating data and visual display (Fig. 6.15, Fig. 6.16), the manager immediately looks at the data from different sides that may (or may not) be related. with a viral problem.

Next, we create different indicators of business among ourselves, trying to reveal the presence of mutual connections; You can look at the data more carefully, detailing them, for example, by distributing them in warehouses by hour, by region or by clients, or, for example, further refine the information provided to obtain more details i. Please follow the additional module statistical estimation and simulation modeling There will be a number of options for the development of this approach, and from them the most pleasant option is selected.

Rice. 6.15.

The firm's ceramics specialist, for example, may formulate a hypothesis that the growth of assets in different branches of the enterprise will lie due to the relationship between certain facists with technical and economic education. To verify this hypothesis, the manager can request these comparisons and display the correlation for those filials for which the growth of assets in the current quarter has decreased at the same level as in the past. 10%, and for those who have moved more than 25 %. It is your fault that your mother may be able to resist a simple choice from the presented menu. As soon as the results are obtained, they will clearly fall into two distinct groups, which may serve as a stimulus for further re-examination of the proposed hypothesis.

In this hour, a quick development has been taken away directly, titles dynamic modeling(Dynamic Simulation), which increasingly implements the values of the larger FASMI principle.

Using dynamic modeling, the analyst will be a model of the business situation that develops hourly, following a given scenario. In this case, the result of such modeling may be several new business situations that will generate a tree of possible solutions based on an assessment of the marketability and prospects of the skin.

Rice. 6.16.

Table 6.3 shows the equal characteristics of static and dynamic analysis.

Table 6.3.

Characteristic	Static analysis	Dynamic analysis
Tipi power	Who? What? How much? Yak? If? De?	Why so? What would happen, what...? What will happen...?
Airtime hour	Not regulated	Seconds
Typical operations with data	Regulations sound, diagram, table, little picture	Sequence of interactive sounds, diagrams, screen forms. Dynamic change of peer aggregation and data cross-section
The range of analytical benefits	Middle	High
Type of screen forms	Mainly due to regulations	Refers to koristuvach, and the possibility of customization
Data aggregation range	Detailed and vague	Appears to be a koristuvach
"Vik" data	Historical and accurate	Historical, accurate and forecasted
Types of queries	Basically, retrained	Neperedbachuvani - hour after hour
Assignment	Analytical processing is regulated	Extensive analysis, modeling and forecasting

It is almost always necessary to set up a routine analytical system for a wide range of data analysis – this is a required task A single, fully functional information system based on heterogeneous software features and solutions. And the choice of costs for the implementation of the IS itself becomes a very difficult task. There are a number of factors involved here, including the mutual influence of various software components, ease of their mastery, quickness and integration, efficiency of operation, stability and shape, the similarity and potential of mutual relations between different manufacturers.

OLAP is based on the task of analyzing multifactorial data. However, given any table with data that requires one descriptive column and one column of numbers, an OLAP tool will be an effective way to analyze and generate feedback. As an example of the application of OLAP technology, we will look at the results of the sales process.

Key nutrition: “How much was sold?”, “How much was sold?” the complexity of business and the accumulation of historical data are expanding in the world to a large number of factors, or divisions: ".. in St. Petersburg, in Moscow, in the Urals, in Siberia...", ".. in the past quarter, in the ishnim" , "..from postal owner A versus postal owner B..." then.

Evidence for such food is necessary for making management decisions: about changing the assortment, prices, closing and opening stores, branches, terminating and signing agreements with dealers, conducting or launching advertising campaigns y then.

If you try to see the main figures (facts) and views (arguments of data) that the analyst manipulates, trying to expand and optimize the business of the company, then you will get a table suitable for analyzing sales as a template, which This is a specific skin treatment for a specific company.

Hour. As a rule, there are a number of periods: River, Quarter, Month, Decade, Week, Day. Many OLAP tools automatically calculate older periods from a date and calculate results for them.

Product category. Category may be splints, stinks vary for the skin type of business: Variety, Model, Type of packaging, etc. If only one product is sold or the assortment is very small, then the category is not required

Product. Sometimes the name of the product (or service), its code or article number are reviewed. In these cases, if the assortment is very large (and businesses have tens of thousands of positions in their price list), a basic analysis of all types of goods may not be carried out, but narrowed down to certain convenient categories.

Region. In a global business, one can take into account the Continent, Group of countries, Country, Territory, Place, District, Street, Part of the street. Of course, since there is only one retail outlet, the whole world is worldwide.

Seller. This world also depends on the structure and scale of the business. Here you can be: Branch, Store, Dealer, Sales Manager. In some cases, the world disappears every day, for example, if the seller does not pay attention to his duties, there is only one store, and so on.

Buyer. In some cases, for example, in general trading, the buyer has no information and dies every day, in other cases, information about the buyer is important for the sale. This example can be named after the purchasing company or without grouping the characteristics of the clients: Galuz, Group of Enterprises, Vlasnik, and so on. Analysis of the sales structure to identify the most important warehouses in the market to select. For this purpose, it is important to use, for example, a diagram of the “Pirig” type in folding situations, if you are looking at section 3 of the series - “Stovpts”. For example, the “Computer Equipment” store collected $100,000 in computer sales during the quarter, $10,000 in photographic equipment, and $4,500 in commercial materials. Abstract: the store's turnover is largely due to the sale of computers (in fact, it is possible that consumable materials are necessary for the sale of computers, and also the analysis of internal deposits).

Dynamics analysis ( regression analysis- Revealing trends). Revealing trends, seasonal changes. Initially, the dynamics are displayed by a graph of the “Line” type. For example, sales of Intel products have been declining over time, while Microsoft's sales have been increasing. It is possible that the average buyer will become more kind, or the image of the store, and with it the warehouse of buyers, will change. It is necessary to adjust the assortment. Another example: over a period of 3 years, the reduction in sales of video cameras decreases.

Analysis of deposits(Correlation analysis). Renewal of obligations for the sale of various goods in the hour to identify the necessary assortment - “basket”. For this purpose, you should manually create a graph of the “Line” type. For example, while looking at the range of printers over the first two months, there was a drop in sales of powder cartridges.

Main features of OLAP technology (Basic):

there is a rich variety of conceptual representations of these data;
intuitive manipulation of data;
availability and detail of data;
package of data versus interpretation;
OLAP analysis models;
"client-server" architecture (OLAP accessible from the desktop);
insight (insightful access to external data);
insured for a rich supply of support.

Special features:

processing of informal data;
saving OLAP results: saving them separately as output data;
turning off daily values;
processing of daily values.

Features of the reporting (Report):

flexibility of molding;
standard productivity of people;
automatic adjustment of the physical level of data acquisition.

Dimension Management:

the universality of extinctions;
the amount of vibration and similar aggregation cannot be determined;
The number of operations between dimensions is not limited.

Historically, it has turned out that today's term "OLAP" is respected not only by the rich view of the data from the side of the end user, but also by the rich representation of data in the entire database. This is also related to the appearance of the independent terms “Relational OLAP” (ROLAP) and “Multiple OLAP” (MOLAP).

OLAP service is a tool for analyzing large amounts of data in real time. Interacting with the OLAP system, you can perform a detailed review of information, extract additional data and perform analytical operations such as detailing, sampling, cross-section Ilu, changes in hours overnight due to a variety of parameters. All work with the OLAP system is carried out in terms of the subject area and allows for a statistically based model of the business situation.

OLAP software is a tool for the rapid analysis of data stored in storage. The main feature is that they are oriented towards the development of not a specialist in information technology, not an expert statistician, but a professional in applied management - a manager of a department, a department, She will be named director. Kostya is intended to connect the analyst with the problem, and not with the computer. In Fig. 6.14 readings is an elementary OLAP cube that allows you to evaluate data in three dimensions.

A rich OLAP cube and a system of advanced mathematical algorithms for statistical processing allow you to analyze data of any complexity at any time intervals.

Rice. 6.14. Elementary OLAP cube

Next, we create different indicators of business among ourselves, trying to reveal the presence of mutual connections; You can look at the data more carefully, detailing them, for example, by distributing them in warehouses by hour, by region or by clients, or, for example, further refine the information provided to obtain more details i. Then, with the help of an additional module of statistical estimation and simulation modeling, there will be a number of options for the development of the approach, and from them the most suitable option will be selected.

Rice. 6.15.

At this time, a rapid development has been taken away directly from the title of dynamic modeling (Dynamic Simulation), which increasingly implements the meaning of the greater FASMI principle.

Rice. 6.16. Analytical IS processing, data processing and information submission

Table 6.3 shows the equal characteristics of static and dynamic analysis.

Purpose of the testimony

This evidence is about one of the categories of intelligent technologies, which are manual analytical tools - OLAP technologies.

Meta evidence: open and highlight 2 points: 1) the concept of OLAP and its practical significance in financial management; 2) implementation of OLAP functionality in software solutions: strengths, capabilities, advantages, shortcomings.

I would like to point out that OLAP is a universal tool that can be used in any application, and not only in finance (as it may be clear from the name of the evidence), which will require analysis of data Using other methods.

Financial management

Financial management is an area where there is no other important analysis. Whether financial and management decisions are made is the result of previous analytical procedures. Today, financial management is assuming a significant role for the successful functioning of a business. Regardless of the fact that financial management is an additional process in the enterprise, it will require special respect, as some of the financial and management decisions can lead to great losses.

Financial management is aimed directly at ensuring the receipt of financial resources in the necessary areas, at the right time and in the right place, with the aim of achieving the maximum effect from them by ensuring optimal distribution.

Perhaps it is important to determine the level of maximum efficiency of resource utilization, and in any case,

The financial director is first and foremost guilty of the nobility:

How many financial resources are there?
Do stars have to pay money and in what obligations?
where to invest more effectively and why?
And at what times do you still need to work?
How much is required to ensure normal business operations?

In order to identify evidence based on nutrition, it is necessary to analyze and know how to analyze a large number of indicators of activity. In addition, FU covers a large number of areas: analysis of penny flows, analysis of assets and liabilities, profitability analysis, margin analysis, profitability analysis, assortment analysis.

Zannanya

Therefore, the key factor in the effectiveness of the financial management process is the clarity of knowledge:

Peculiarities of knowledge of the subject matter (one might say theoretical and methodological), including evidence, intuition of a financier/financial director
Underground (corporate) knowledge or systematized information about the facts of financial transactions in the enterprise (information about the past, current and future enterprises, is presented in various shows nikah and vimirah)

If it is primarily in the sphere of activity of a particular financier (or a director of personnel who hires a particular employee), then the other may be directly involved in the work of specialists in financial and information services.

What is it now

Nowadays there is a typical paradoxical situation in enterprises: there is already too much information, too much. But she is in a chaotic state: unstructured, unorganized, disjointed, not always reliable and often amended, it is practically impossible to know and remove. Trinavla was taken by the one, Marne General GiR Fіnansovo Zvitnosti, Yaka is unprotected for the Fіnansky analiza, Vadka of the Popriyatty, Sailing Star of the Selce of Office, and the Nadannya Control authorities.

The results of the investigation conducted by the company Reuters Among 1,300 international managers, 38% of them say they spend a lot of time trying to find the information they need. It turns out that highly qualified fakhvets spend high-paying hours not on data analysis, but on collecting, searching and systematizing the information necessary for this analysis. At the same time, managers understand the importance of engaging with data, as they often fail to pay close attention to the task, which again reduces the effectiveness of their work. The reason for this situation: too much information and lack of knowledge.

What do you need to be robotic?

Information can be transformed into knowledge. For daily business, information is valuable, its systematic acquisition, synthesis, exchange, exchange is a kind of currency, but in order to extract it, it is necessary to maintain information as a business process .

The key to information management is the delivery of necessary information to the right people within the organization at a specific time. The aim of such management is to help people to better manage the growing amount of information at once.

Information technologies in this regard act as a means by which it would be possible to systematize information in the enterprise, give existing employees access to it and the tools to transform the value Information for knowledge.

Basic concepts of OLAP technologies

OLAP technologies (On-Line Analytical Processing) is not the name of a specific product, but a whole technology for the rapid analysis of a wealth of data accumulated in the system. To understand the essence of OLAP, it is necessary to look at the traditional process of sifting through information to reach a decision.

The traditional support system will be solved

Here, of course, there may be a lot of options: new information chaos or the most typical situation, if the enterprise has operating systems, in addition to which the facts of the current operations are recorded and their storage in databases. In order to extract data from databases for analytical purposes, a system of querying data samples was created.

If this method of support is used to reduce the flexibility, it may have a lot of shortcomings:

A very small number of data are analyzed that may be useful for making a decision
Sometimes complex, rich information is created, from which 1-2 rows are actually highlighted (otherwise - about any kind of fallout) - informational importance
There is a perfect reaction to the process of change: if a new supply of data is necessary, then it requires formal descriptions and coding by the programmer, and then later writings. Hour of recovery: years, days. And perhaps a decision is urgently needed, for sure. Even after receiving new information, new food (to be specified)

If the calls for queries are submitted in a single-world format, then the business problems will be wide-ranging and wide-ranging. If you need to get a clear picture of the company's business, it is necessary to analyze the data in different sections.

Many companies create excellent relational databases, ideally sorting out mountains of information, which in and of themselves will not ensure a reliable response to market conditions. SO – relational databases will be the most suitable technology for saving corporate data. We are not talking about a new database technology, but rather about instrumental analysis tools that will complement the functions of existing DBMSs and add features to transfer and automate various types of intelligent analysis, OLAP .

Understanding OLAP

What does OLAP do?

Blame tools for access to data collection
Dynamic interactive data manipulation (wrapping, consolidation or detailing)
Initially visual representation of data
Fluidity – analysis is carried out in real time
Worldwide data supply - one-hour analysis of low indicators from many worlds

To eliminate the effect of the use of OLAP technologies, it is necessary to: 1) understand the essence of the technologies themselves and their capabilities; 2) clearly identify which processes need to be analyzed, which indicators of stench will be characterized and in which worlds they must be thoroughly studied in order to create a model for analysis.

The basic concepts that OLAP technologies operate with are:

Rich worldliness

To understand the richness of the data, first of all, submit a table that displays, for example, the cost of a business with economical elements and a business unit.

This data is presented in two worlds:

statya
business unit

This table is not informative, but only shows sales for one period of time. For different hour periods, analysts will have to create a table (for each hour period):

The baby can see the 3rd Vimir, Hour, in addition to the first two. (Stattya, business unit)

Another way to show a wealth of data is to show it in a cube shape:

OLAP cubes allow analysts to sift through data across multiple slices to generate nutritional and business decisions:

What expenses are critical for which business units?
How can I spend a business unit per hour?
How do the income tax statistics change per hour?

Evidence for such nutrition is necessary for making management decisions: about the shortening of old expenditure items, influx into its structure, identifying the reasons for changes in expenditures in the hour, updating the plan and their elimination - optimization tion of its structure.

Whose butt has more than 3 dimensions examined. It is important to represent more than three worlds, but it works the same way as in the case of three worlds.

If you use OLAP add-ons, you can extract data for 3 or more views, for example, you can add one more view – Plan-Fact, Income category: direct, indirect, Agreement, Months. Additional vimirvaniya allows you to isolate more analytical samples and provide nutritional supplements with more minds.

Hierarchy

OLAP also allows analysts to organize their data into a hierarchy, which consists of groups and subgroups of subgroups of values that display indicators across the entire organization - the most logical way To analyze the business.

For example, you can completely group it hierarchically:

OLAP allows analysts to take data from a single metric (at the topmost level), and then drill down to the bottom and next level, thereby revealing the exact reason for changing the metric.

By allowing analysts to edit a small amount of data that may be subject to hierarchical urges, OLAP allows you to capture the picture of a business that is not constrained by the structure of the information shed.

Changing the direction of analysis in Cuba (data wrapper)

As a rule, we operate with the following concepts: values, specified in columns, rows (these can be a number), other forms of sections, instead of a table, form dimensions (sales, expenses, costs)

As a rule, OLAP allows you to change the orientation of the cube, thereby presenting the data in various manifestations.

The image of the cube data is stored in:

orientation of the vimirs: what vimirs are set in rows, columns, sections;
groups of demonstrators seen in rows, stands, sections.
The change of vimiryuvan lies with the deeds of the koristuvach.

Thus, OLAP allows you to carry out various types of analysis and understand their relationships between their results.

Analysis of illness is an analysis according to the plan, which is complemented by a factor analysis of the causes of illness and detailed indicators.
Analysis of positions: OLAP allows you to identify different positions between different changes, for example, when looking at the assortment of beer over the first two months, a drop in sales was observed.
Establishment (periodical analysis). The consistency of the results of changing the indicator hourly, for a given group of products, in different regions and others.
Analysis of dynamics makes it possible to identify specific trends in changes in indicators over the hour.

Efficiency: we can say that OLAP is based on the laws of psychology: the ability to process information queries in real time - at the pace of the process of analytical interpretation of the data by the correspondent.

Since a relational database can retrieve close to 200 records per second and write 20, then a good OLAP server, vikoryst rows and stacks, can consolidate 20,000-30,000 records (equivalent to one record in a relational database) in a second.

Initiality: It should be noted that OLAP allows for the development of the graphical presentation of data from the end user. The human brain is able to perceive and analyze information presented in the form of geometric images, in view of many orders of magnitude more information presented in alphanumeric form. Butt: Please don’t forget that you need to know the identity of one of the hundred photographs. I respect that this process will take you a little more than a little time. And now realize that instead of photographs you will be given a hundred verbal descriptions of these people themselves. I think that you will not be able to verify the assigned task.

Simplicity: The main feature of these technologies is that they are oriented not by a specialist in information technologies, not by an expert statistician, but by a professional in applied research - a manager of a credit department, a manager of a budgetary nareshti, director. They are intended to focus the analysis on the problem, not on the computer..

Regardless of the great capabilities of OLAP (besides, the idea is quite old - the 60s), the reality is practically not stagnant in our enterprises. Why?

Daily information or not possible
Zvichka think twice
price bar'er
The technological level of articles dedicated to OLAP is remarkable: to identify unimportant terms – OLAP, “data mining”, “unregulated queries”, “detection of historical correlations”

Our approach is the best way to stop OLAP

In addition, we also have a specific understanding of the application value of OLAP based on reasonable technological capabilities.

Our Russian authors of various materials devoted to OLAP express the following idea due to the complexity of OLAP: OLAP is more widely understood as a tool that allows you to dig up and digest data simply and manually, strange manipulations that come to the analyst’s mind during the analysis process. The more “views” and “views” these analysts have, the more ideas they have, which, in their own way, generate new and new “views” for verification. This is not correct.

The approach to understanding the value of OLAP is based on a methodological analysis model, which requires an OLAP solution to be developed when designing. The analyst is not guilty of playing with the OLAP cube and aimlessly changing its data, detailing, orientation of data, graphical display of data (and this is useful!), and clearly understand how and identifying your needs, in what order (primarily, the elements of critical "There may be buti here, but this is not the main element of OLAP functionality).

Applied OLAP Wikistan

Budget
Rukh of penny koshtivs

One of the most beneficial problems is the use of OLAP technology. It is not for nothing that the daily budgeting system is not considered complete without the availability of OLAP tools for budget analysis. Most budget projects can easily be based on OLAP systems. In this case, the data suggests a wide range of nutrition: analysis of the structure of expenditures and income, the level of expenditures for financial statements in various departments, analysis of the dynamics and trends of expenditures for financial statements, analysis of Costs and profits.

OLAP allows you to analyze income and cash flows across business operations, counterparties, currencies, and time using the method of optimizing their flows.

Financial and management information (with analytics required by management)
Marketing
Balanced Scorecard
Analysis of profitability

For the availability of similar data, you can use different OLAP technology programs.

OLAP products

In this section we will talk about OLAP and the software solution.

Additional benefits to OLAP products

There are many ways to implement OLAP add-ons, then each specific technology is not obligatory, but rather recommended. For different minds and circumstances, one approach may be worth another. The implementation technique includes a lot of different patented ideas, which are what the authors write: different types of “client-server” architecture, analysis of time series, object orientation, optimization of saving them, parallel processes then. However, these technologies cannot be part of the value of OLAP.

Є characteristics that can be observed in all OLAP-products (including OLAP-products), which are the ideal technology. There are 5 key values that characterize OLAP (as called the FASMI test): Swedish Analysis of Separated Biogeometric Information.

Shvidky(FAST) – means that the system can ensure that the majority of responses are visible to traders within approximately five seconds. If the system is ahead of the curve, so that the process is more thorough, students can become overwhelmed and lose their thoughts, in which the strength of the analysis suffers. Such flexibility is not easy to achieve with a large number of data, especially since special calculations are required “for the lot”. Customers go to a wide range of methods to achieve these goals, including specialized forms of data storage, large forward calculations, and powerful hardware benefits. There are no further optimized solutions available today. At first glance, it may seem surprising that when the price of the hulk has been taken away, as the days have recently arrived, the koristuvach is already beginning to feel the need to recover, and the project appears much less successful, different types of responses, at the cost of less detailed analysis.
Divided means that the system allows you to save all possible data and implement divisions and instant access to data for different levels of accounts. The system is responsible for processing numerical data changes in its own, secure way. This is the main weakness of many OLAP products, which tend to assume that all OLAP add-ons require less reading and provide simpler security.
Richly worldly- Klyuchova vimoga. If it was necessary to define OLAP in one word, we would choose yogo. The system can provide a rich conceptual presentation of data, including continued support for hierarchies and multiple hierarchies, while providing the most logical way to analyze a business. The minimum amount of data that may be generated is not installed, but remains in the programs, and most OLAP products have a sufficient amount of data for these markets, such as they are aimed. And again, it is not clear that the underlying technology of the data base is subject to corruption, since the data base effectively extracts a wealth of conceptual information. This particularity is the core of OLAP
Information. Necessary information may be discarded where it is necessary, regardless of the obligation to save it. However, there is a lot left behind the programs. The weight of various products is measured in terms of how many inputs can be processed and how many gigabytes of waste can be saved. The complexity of the products is already clear - the largest OLAP products can process thousands of times a large amount of data from those on par with the smaller ones. This drive requires a variety of factors, including data duplication, required RAM, disk space, performance indicators, integration with information storage devices, etc.
Analysis This means that the system can cope with any logical and statistical analysis characteristic of a given program, and will ensure that savings are visible to the end user. Koristuvach is responsible for the ability to introduce new special calculations as part of the analysis without the need for programming. Thus, all the necessary functional capabilities of the analysis must be provided in an intuitive way for end users. The analysis could include song procedures, such as analysis of hourly series, division of expenses, currency transfers, search for purposes, etc. Such possibilities vary widely among products, depending on the target orientation.

In other words, these are 5 key values – the goals of achieving any OLAP-oriented products.

Technological aspects of OLAP

The OLAP system includes song components. Find out different schemes of their work, which other products can be sold.

Components of OLAP systems (what makes up an OLAP system?)

Typically, an OLAP system includes the following components:

Dzherelo danikh
Dzherelo, from whom the data for analysis is taken (data shed, database of operational cloud systems, set table, combinations of overhauled).
OLAP server
The data is transferred or copied to the OLAP server, then systematized and prepared for subsequent generation of query responses.
OLAP client
Interface of Koristuvach to the OLAP server in which Koristuvach operates

Please note that not all components are binding. There will be desktop OLAP systems that allow you to analyze data that is stored directly on the user’s computer and does not require an OLAP server.

Prote is a binding element that is important for food: the presence of food is more important. It looks like an Excel table, a cloud system database, or a structured branch of an IT file can be integrated with an OLAP system directly or indirectly creations. For which OLAP systems special tools are created. Since there is no such data, because the stench of insufficient strength and virulence lingers, OLAP will not help. But OLAP is the only thing that overdos the data, and if there is no data, the stench becomes like a dirty river.

Most data for OLAP programs resides in other systems. However, in certain programs (for example, for planning or budgeting), data can be created directly in OLAP programs. If the data comes from other programs, make sure that the data is saved in a separate, duplicate form for the OLAP program. That’s why it’s completely necessary to create a bunch of data.

Please note that the term OLAP is closely related to the term Data Warehouse. The collection of data is subject-oriented, time-bound and constant collection of data to support the decision-making process. Data is typically collected from operating systems (OLTP systems) designed to automate business processes, which can be combined with other external tools, such as statistical data.

Regardless of those who place apparently supernatural information, such as in databases or files of operating systems, the similarity of data is necessary because:

fragmentation of data, saving it in different DBMS formats;
The productivity of data extraction is improving
Since all data is stored on a central database server (which is extremely rare), the analyst cannot understand their complex, sometimes confusing structures.
Complex analytical queries to operational information balance the company's flow, permanently blocking tables and consuming server resources
Possibility of cleaning and improving data
Analyzing data from operating systems is neither possible nor very important;

Preservation of the conjunctiva - add “syrovin” for analysis in one place and in a simple, sensible structure. This concept of Data Cache is a concept for analyzing data, rather it is a concept for preparing data for analysis. Vaughn transfers the implementation of a single integrated data engine.

OLAP products: architectures

When choosing OLAP products, 2 foods are important: how and where saveі earn money tribute It is important to understand how the two processes are implemented and separate OLAP architectures. There are 3 ways to save data for OLAP and 3 ways to process this data. There are plenty of researchers who have put forward a number of options that try to convey that their approach is the only reasonable one. This, of course, is absurd. However, very few products can be processed more clearly in one mode.

Options for saving OLAP data

Saving in this context means the replacement of data in a station that is constantly being updated.

Relational databases: this is a typical choice, since cloud data is stored in the RDB. In most cases, the data is saved in a denormalized structure (the most useful “star” scheme). A normalized database is not acceptable due to the very low productivity of the queries when forming aggregated values for OLAP (often aggregated data is stored in aggregate tables).
Database files on the client computer (kiosks or data windows): this data can be later distributed or created behind the power of client computers.

Global Databases: It is assumed that the data is stored in a global database on the server. You can include data pulled from and compiled from other systems, relational databases, end-user files, etc. In most products, a large number of data bases are stored on disk, and some products allow the storage of data and random access memory, which accounts for the most frequently used data on the lot.” Even in a small number of products based on a large number of data bases, it is possible to edit the data multiple times, many products allow a single change, but multiple readings of the data, while others are interchanged with more than one reader.

These three days of saving data vary depending on the possibilities of saving, and they are sorted out in order, which decreases according to the possibilities. There are also different characteristics of productivity when implementing queries: relational databases work much better, but there are two other options.

Options for processing OLAP data

There are the same 3 options for processing data:

SQL Victorial: This option is especially useful when storing data in the RDB. However, SQL does not allow you to do a lot of calculations in a single statement, so you need to write complex SQL statements in order to achieve less than a lot of functionality. However, this does not mean that the distributors do not carry out the tests. In most cases, it is necessary to combine a number of similar calculations in SQL with the results that can be obtained from a large number of data processing or from client machines. It is also possible to use RAM, which can save data that is used in more than one entry: this has radically improved the output.
There is a lot of processing on the client: the client OLAP product does this calculation on its own, but such processing is only available in the context of the fact that users are constantly running their PCs.

Rich processing on the server: this is a popular place for creating rich calculations in client-server OLAP applications that are used in many products. Productivity is high, because most of the calculations have already been made. However, this will require a lot of disk space.

Matrix of OLAP architectures

Obviously, to identify saving/processing options, you can select the matrix of OLAP system architectures. Apparently, theoretically, 9 different methods can be found. However, if 3 of them have been removed, then in reality there are only 6 options for saving and processing OLAP data.

Options for saving the world's wealth tributes Options rich in world Tributes	Relational database	Servers, rich global database	Client computer
	Cartesis Magnitude
Rich world server build	Crystal Holos (ROLAP mode) IBM DB2 OLAP Server CA EUREKA:Strategy Informix MetaCube Speedware Media/MR Microsoft Analysis Services Oracle Express (ROLAP mode) Pilot Analysis Server Applix iTM1	Crystal Holos Comshare Decision Hyperion Essbase Oracle Express Speedware Media/M Microsoft Analysis Services PowerPlay Enterprise Server Pilot Analysis Server Applix iTM1
Extensive processing on the client computer	Oracle Discoverer Informix MetaCube	Dimensional Insight Hyperion Enterprise	Cognos PowerPlay Personal Express iTM1 Perspectives

Since the preservation itself means waste, it is customary to group together the options for saving, then:

ROLAP products for sectors 1, 2, 3
Desktop OLAP – in sector 6

MOLAP products – sectors 4 and 5

HOLAP products (allow both global and relational data saving options) – 2 and 4 (shown in italics)

Categories of OLAP products

There are over 40 OLAP clients, although all of them cannot be taken over by competitors, since their capabilities are already being diversified and, in fact, are operating in different market segments. They can be grouped into 4 important categories, which are based on the following concepts: complex functionality - simple functionality, productivity - disk space. It is easy to display the categories in the form of a square, which clearly shows the interconnection between them. The characteristics of rice skin are presented on one side, and the similarities are with others - on the side sides, and those on the back sides are generally good.

Features

Advantages

Nedoliky

Representatives

Application OLAP

Completed add-ons with rich functionality. Everyone may be able to access a rich global data base, but want to work in a relational way. There are a lot of specializations in this category, for example, sales, promotion, banking, budgeting, financial consolidation, sales analysis

Possibility of integration with various programs

High level of functionality

High level of flexibility and scale

Foldability of programs (necessity for basic knowledge)

High quality

Hyperion Solutions

Crystal Decisions

Information Builders

The product is based on a non-relational data structure, which ensures rich storage, processing and presentation of data. Data during the analysis process are selected inclusively from the rich world structure. Regardless of the high level of transparency, buyers are trying to convince buyers to buy their tools.

High productivity (high numbers of total indicators and a wide variety of world transformations, whether due to extinction). The average hour of response to unregulated analytical data in a vicoristan rich-world database is 1-2 orders of magnitude less, lower than the RDB.

High level of transparency: large number of products with which integration is possible

It is easy to cope with the required inclusions to the information model of various functions, carrying out a specialized analysis.

The need for large disk space to save data (due to the multiplicity of the data that is being saved). This extremely ineffective memory storage - due to the denormalization and previously determined aggregation of data from a large global database - results in 2.5-100 times less output detailed data. MOLAP does not allow anyone to work with large databases. The real boundary is a database with a volume of 10-25 gigabytes

Potentiality of the “vibhu” of the data base - uncontrolled, sharp, disproportionate growth of obligations

There is a lot of flexibility for the need to modify data structures. Any change in the structure may soon require a new hypercube.

For a large number of databases, there are currently common interface standards, which describe and manipulate data

Hyperion (Essbase)

DOLAP (Desktop OLAP)

Client OLAP products that are easy to install and get low costs from distribution in one place

We are talking about such analytical processing, where hypercubes are small, their size is small, their needs are modest, and for such analytical processing a personal machine on a desktop is sufficient.

The meta of this market is the automation of hundreds and thousands of workers, and those responsible for performing simple analysis. Buyers are often oriented towards buying more workers, as needed

Good integration with databases: rich, relational

Possibility of making complex purchases, which reduces the cost of supply projects

Simplicity of Wikoristan Programs

Functionality is very limited (cannot be compared to any plan with specialized products)

The tension is very limited (there are few obligations, the amount of vibration is small)

Cognos (PowerPlay)

Business Objects

Crystal Decisions

This is the smallest sector of the market.

Detailed data is lost there, but right away - in the relational database; All aggregates are stored in the same database in specially created service tables

How to do business with very high obligations (economical savings)

They transfer insurance coverage to a wide variety of work modes, including in editing mode, and not just reading

Greater level of data protection and good ability to differentiate access rights

It is possible to make changes as often as possible until the structure is modified (do not require physical reorganization of the database)

Low productivity is significant due to the liquidity of the product (the product on the market is measured in minutes or in years, or even in seconds). These are handy alarm clocks and other interactive analytical tools.

Foldability of products. Raise significant costs for the services of information technology specialists. To ensure the productivity that comes with MOLAP, relational systems require careful processing of database schemas and tuning of indexes, so that great efforts are made on the side of database administrators

Roads for repair

SQL exchanges are devoid of reality, which does not allow the implementation in RDBMS of the rich functions that are easily provided in systems based on a rich data supply.

Information Advantage

Informix (MetaCube)

Please note that hybrid products that allow you to select MOLAP and ROLAP mode, such as Microsoft Analysis Services, OracleExpress, Crystal Holos, IBM DB2 OLAPServer, may also always select MOLAP mode.

The skin from the presented categories has its own strengths and weaknesses, there is no single optimal choice. Vibe influences three important aspects: 1) productivity; 2) disk space for saving data; 3) feasibility, functionality and especially the scalability of the OLAP solution. In this case, it is necessary to take into account the obligations of data, the complexity of technology, the needs of customers and find a compromise between the speed of the data and the excessive disk space occupied by the database, downtime and rich functionality.

Classification of Data Collections is consistent with the target database

Shortcomings of OLAP

As any OLAP technology, it also has its drawbacks: high costs for hardware security, training and knowledge for administrative personnel and end-users, high expenses for the implementation of the project (like pennies, so also time-consuming, intellectual).

Select OLAP product

Choose your OLAP product correctly, but as carefully as you want, so that the project does not fail.

As a matter of fact, the importance of products lies in many areas: functional, architectural, technical. These products are already subject to adjustment. Activities created for specialized subject areas: marketing, sales, finance. Є products for hidden purposes, which do not have a practical application, which may be used with little ones. As a rule, such products are cheaper, less specialized, and then you spend more on sales. The range of OLAP products is very wide - from the simplest functions of tables and diagrams that are included in the warehouse of office products, to the analysis of data and the search for patterns, which can cost tens of thousands of dollars.

As in any other case, in the field of OLAP we cannot have clear recommendations for choosing tools. It is possible to highlight a number of key points and establish the ability of software to meet the needs of the organization. One thing is important: without thinking carefully about how you decide to use OLAP tools, you run the risk of getting yourself a serious headache.

During the selection process, you need to consider 2 elements:

evaluate the needs and feasibility of business
evaluate the current market position, important development trends as well

Then we’ll put everything in order and, finally, make a choice.

Needs assessment

It is impossible to make a rational choice of product without understanding that there is always a risk of abuse. A lot of companies want to eliminate the “greatest virus” without any clear understanding, as they may be victorious.

In order for the project to be successfully implemented, the financial director is required, at a minimum, to correctly formulate his duties and responsibilities to the technician and automation service specialists. A lot of problems arise from insufficient preparation and awareness for the choice of OLAP, IT specialists are aware of the difficulties of understanding even those who manipulate different concepts And the terms and hanging super-literate similarities. There is a need for service from the target within the company.

These factors have already become obvious after taking a look at the OLAP product category, and itself:

Technical aspects

Dzherela data: corporate data warehouse, OLTP system, tabular files, relational databases. The ability to link OLAP tools with the DBMS system used by the organization. As practice shows, the integration of different foods into a stable functioning system is one of the most important nutritional benefits, which can sometimes be associated with great problems. It is necessary to consider how easily and reliably it is possible to integrate OLAP features with those of a DBMS. It is also important to evaluate the possibilities of integration not only with data files, but also with other additions that you may need to export data: email, office programs
Lots of data to take out for insurance
Server platform: NT, Unix, AS/400, Linux - but do not forget to ensure that OLAP products specified by the specification are designed on dubious or dying platforms that you are still victorious about
Client side and browser standards
The architecture is emerging: local network and PC modem connection, high-speed client/server, intranet, extranet, Internet
International features: rich currency support, rich transactions, collective data mining, localization, licensing, Windows updates

Obligations regarding input information, such as that provided to the future

Koristuvachi

Areas of expertise: sales/marketing analysis, budgeting/planning, analysis of performance indicators, analysis of accounting reports, clear analysis, financial analysis, formation of analytical materials (reports)
The number of files and their placement, depending on the level of access rights to these functions, secrecy (confidentiality) of information
View of a professional: higher care, finance, marketing, HR, sales, production, etc.
Dosvid koristuvach. The level of qualification of a business owner. Take a look at the food preparation. It is important that the OLAP client program be such that developers feel confident that they can effectively exploit it.

Key features: need for data recording, division of calculations, complex currency conversion, need for other data, spreadsheet interface, complexity of program logic, required dimensionality, type of statistical analysis: , search for information, analysis of “what is it like”

Vprovadzhennya

Who will be involved in the implementation and operation: external consultants, internal IT service or final correspondents
Budget: software, hardware, services, data transfer. Remember that paying for licenses for an OLAP product is only a small part of the overall benefit to the project. Installation and hardware costs may be higher, but the license fee may be higher, but support, operation and administration costs may be much higher. And if you made the wrong decision to purchase an unsuitable product only because it is cheaper, the rest of you can see a hidden benefit to the project through expenses on maintenance, administration and/or hardware expenses, despite the fact that Otherwise, you are taking away a lower level of business benefits. When you're planning on spending a lot of money, don't forget to consider the following: How wide is the choice of equipment for promotion, training and support? Is the potential reserve fund (service workers, contractors, consultants) growing or shrinking? How widely can you demonstrate your professional qualifications?

Regardless of the fact that the performance of analytical systems today is not high, and the methodology and technology for implementing such systems are still at the stage of their formation, even today the economical effect without taking care of them, the effect really outweighs traditional operating systems.

The effect of proper organization, strategic and operational planning of business development is important to evaluate later in numbers, but it is obvious that tens or even hundreds of times can exceed the costs of implementing such systems. However, don’t have mercy. The effect is ensured not by the system itself, but by the people working with it. Therefore, the declarations on the script are not entirely correct: “the Data Storage system and OLAP technologies help the manager make the right decisions.” Current analytical systems and systems of artificial intelligence and technology can neither help nor hinder the decision-making process. The goal is to provide the manager with all the information necessary to make a decision directly. And what information will be received and what decisions will be made on this basis depends only on a specific person, such as Vikorist.

It goes without saying that systems can help solve a lot of business problems and can have a lasting positive effect. It is impossible to wait only for those who are the first to recognize the advantages of this approach and appear before others.

4. Classification of OLAP products.

5. Principles of OLAP clients.

7. Areas of development of OLAP technologies.

8. Application of using OLAP technologies for analysis in the sales area.

1. The place of OLAP in the information structure of a business.

The term "OLAP" is closely related to the term "Data Warehouse".

Data is extensively collected from operating systems (OLTP systems), which are used for automating business processes. In addition, the sensation may be influenced by the influence of external sources, for example, statistical data.

Preservation of the conjunctiva - give “syrovin” for analysis in one place and in a simple, sensible structure.

Another reason that plausibly leads to the emergence of a hidden analogy is that complex analytical queries to operational information interfere with the company’s flow of work, permanently blocking tables and wasteful server resources.

Under the hood, you can understand the not necessarily gigantic accumulation of data - smut, so that it is handy for analysis.

Centralization and manual structure are not all that an analyst needs. You also need a tool for reviewing and visualizing information. Traditional sounds were created on the basis of a single synchrony, eliminating one thing - humility. They cannot be “twisted”, “burned up” or “burned up”, in order to remove the meaning of the data presented. I need such a tool that allows me to gorge and gorget tributes simply and manually! OLAP is such a tool.

Although OLAP is not a necessary attribute of a data warehouse, it is increasingly necessary to analyze the data accumulated in a data warehouse.

The place of OLAP in the information structure of a business (Fig. 1).

Malyunok 1. MіstseOLAP in the information structure of the enterprise

Operational data is collected from various sources, cleaned, integrated and compiled into a relational system. In this case, the stench is already available for analysis for additional different types of calls. Further (generally or partially) preparations are made for OLAP analysis. They may be included in a special OLAP database or absent from a relational analogy. The most important element is metadata, which is information about the structure, placement and transformation of data. This will ensure effective interaction between the various components of the conduit.

In other words, OLAP can be defined as a set of methods for the wide-ranging analysis of data accumulated in storage.

2. Operational analytical compilation of data.

The OLAP concept is based on the principle of a rich and diverse representation of data. In 1993, EF Codd looked at the few parts of the relational model, first of all, pointing out the impossibility of “combining, reviewing and analyzing data from the point of view of multiplicity of changes, so that it would be most useful for corporate analysis.” in a ticking way", and having identified the hidden capabilities to OLAP systems that will be expanded The functionality of relational DBMSs includes rich analysis as one of its characteristics.

According to Codd, a multi-dimensional conceptual view of data is a multi-dimensional perspective that consists of several independent dimensions, which can be analyzed from the totality of data.

A one-hour analysis of many worlds is considered a rich world analysis. Kozhen vimir includes directly the consolidation of data that is formed from a series of successive levels of analysis, where the current one is largely consistent with the aggregation of data for the same Iruvannyam.

Thus, the death of Vikonavets can be seen as a direct consolidation, which is formed from the equalization of “enterprise - subdivision - division - service”. The changing hour can now include two direct consolidations - “river - quarter - month - day” and “week - day”, as the ratio of hours to months and even more unreasonable. In this case, it becomes possible to select the desired level of detail of information for skin treatment.

The drilling down operation represents a collapse from higher stages of consolidation to lower ones; However, the rolling up operation means a move from lower levels to higher ones (Fig. 2).

Figure 2.Vibration and direct consolidation of data

3. Access to the capabilities of operational analytical processing.

The rich approach to wine is practically overnight and in parallel with the relational one. However, only starting from the mid-90s, and more precisely from
born 1993, interest until MDBMS having begun to develop a zagal character. A new program of one of the founders of the relational approach has appeared e. kodda, in which we formulated 12 main benefits before implementation OLAP(Table 1).

Table 1.

	There is a wide variety of data available worldwide	Koshti may support a richly world-wide view of the data on a conceptual level.
	Serendipity	The correspondent is not obliged to know what specific methods are used to save and process data, how data is organized and evidence is taken.
	Availability	It is your responsibility to choose the ones yourself and contact the best one to formulate a type of data. Cats are responsible for ensuring the automatic generation of their own logical circuits in a variety of heterogeneous data sources.
	Improved productivity	Productivity practically does not have to lie in the amount of waste in your drink.
	Support for client-server architecture	Koshti may work in client-server architecture.
	Equality of all worlds	The origin of the extinctions may be basic, all of them may be equal (symmetrical).
	Dynamic processing of rarefied matrices	Unimportant responsibilities must be avoided and dealt with in the most effective way.
	Support for the richly insured regime of work with data	It is the responsibility of the cats to ensure the ability to work with more than one person.
	Support for operations based on different worlds	All world-wide operations (for example, Aggregation) are responsible for the same time and are expected to stagnate until a certain number of extinctions occur.
	Ease of data manipulation	Koshty guilty mothers have the most user-friendly, natural and comfortable interface for the user.
	We apologize for the presentation of data	Kostya is responsible for promoting different ways of visualizing (submitting) data.
	It is not possible to limit the amount of data aggregation and data aggregation	It is not wrong to limit yourself to the number of extinctions that you support.

Rules for evaluating software products for the OLAP class

A set of these made it possible, which became actual OLAP values, to be considered as recommendations, and specific products to be evaluated step by step until they are ideally consistent with all possibilities.

Codd's later development was developed into the so-called FASMI test, which is intended to provide an OLAP add-on, providing the ability to quickly analyze the wealth of information that is shared.

Remembering Codd's 12 rules is very difficult for most people. It turned out that you can summarize OLAP-values with just five keywords: Swidy Analysis of separate rich atomic information - or, briefly - FASMI (translation from English:F ast A analysis of S hared M ultradimensional I information).

This significance was first formulated in early 1995 and has not required any revision since then.

FAST ( Shvidky - means that the system must ensure that most of the evidence is shown to the clients in about five seconds. In the simplest case, requests are processed within one second or even less - up to 20 seconds. Research has shown that the end-users perceive the process immediately, as the results are not removed after 30 seconds.

At first glance, it may seem surprising that when the price of the hulk has been taken away, as the days have recently arrived, the koristuvach is already beginning to feel the need to recover, and the project appears much less successful, different types of responses, at the cost of less detailed analysis.

ANALYSIS (Analysis)This means that the system can cope with any logical and statistical analysis characteristic of a given program, and will ensure that savings are visible to the end user.

It is not so important that this analysis is based on the official tools of the postmaster or on a related external software product such as a spreadsheet, simply all the functional capabilities of the analysis that are necessary are not printed in an intuitive way for end-users. The analysis may include detailed procedures, such as analysis of time series, division of funds, currency transfers, search for targets, changes in rich world structures, non-procedural modeling, identification of blame situations, data mining and other operations. tsії, deposits as an additional supplement. Such possibilities vary widely among products depending on the target orientation.

SHARED means that the system implements all possible privacy protections (possibly up to the middle level) and if multiple write access is necessary, it will ensure blocking of modifications on a specific level. Not all programs require a return data recording. However, the number of such programs is growing, and the system must handle many modifications in its own, secure way.

MULTIDIMENSIONAL (Bagatomirny) - tse key vimoga. If you needed to define OLAP in one word, choose yogo. The system must provide a rich conceptual representation of data, including continued support for hierarchies and multiple hierarchies, as the most logical way to analyze the business of an organization. There is no minimum amount of mitigation that may occur, the fragments will also remain in the programs, and most OLAP products have a sufficient amount of evaporation for these markets, such as ієнтуні.

INFORMATION - that's all. Necessary information may be removed where it is needed. However, there is a lot left behind the programs. The weight of various products is measured in terms of how many inputs can be processed and how many gigabytes of waste can be saved. The complexity of the products is already clear - the largest OLAP products can process thousands of times a large amount of data from those on par with the smaller ones. This drive requires a variety of factors, including data duplication, required RAM, disk space, performance indicators, integration with information storage devices, etc.

The FASMI test is a reasonable and reasonable determination of goals for achieving any OLAP orientation.

4. ClassificationOLAP-Products

Well, the essence of OLAP It is believed that the information available for analysis is presented in the form of a richly dimensional cube, and it is possible to sufficiently manipulate it and remove the necessary information sections - iti. At the same time, the cube is used as a rich dynamic table that automatically sums up data (facts) from different sections (dimming), and allows interactive calculations with calculations and form of information. These operations will be ensured OLAP -by car (or by car OLAP calculation).

Today, the world has a wide variety of products that can be sold OLAP -technology. To make it easier to navigate among them and to classify the classifications OLAP -products: the method of saving data for analysis and the place of acquisition OLAP -Machine. Let's take a closer look at the skin category OLAP products.

Classification according to the method of saving data

Rich cubes will be based on output and aggregate data. Output and aggregate data for cubes can be stored in both relational and global databases. Therefore, at this time, there are three ways to save money: MOLAP (Multidimensional OLAP), ROLAP (Relational OLAP) and HOLAP (Hybrid OLAP) ). Apparently, OLAP - products based on the method of storing data are divided into three similar categories:

1. At the time MOLAP , output and aggregate data are saved in a multi-world database or in a multi-world local cube.

2. At ROLAP - products output data is stored in relational databases or flat local tables on a file server. Aggregate data can be placed in a service table in the same database. The transformation of data from a relational database into a large number of cubes is performed per query OLAP-koshti.

3. U raz vikoristannya HOLAP architecture, the output data is removed on a relational basis, and the aggregates are located in a rich world. Pobudova OLAP - the cube is followed by a capitalization OLAP -Koshti based on relational and rich global data.

Classification by location OLAP-Machine.

For this sign OLAP -products are divided into OLAP servers and OLAP clients:

· For server OLAP - the methods of calculating and saving aggregate data are determined by the same process - the server. The client program takes the results of queries up to a large number of cubes, which are saved on the server. Deyaks OLAP -Servers support the storage of data only in relational databases, and only in large-scale ones. A lot of today's OLAP -Servers support all three ways of saving data:MOLAP, ROLAP and HOLAP.

MOLAP.

MOLAP - tse Multidimensional On-Line Analytical Processing, This is a rich OLAP.This means that the data storage server contains a wide-ranging database (WDB). Sens vikoristannya MBD is obvious. You can effectively save a lot of money on your data by ensuring smooth servicing of requests to the database. The data is transferred via a data engine to a large database, and then the database is subject to aggregation. The forward layout is the one that speeds up OLAP queries; a portion of the data has already been compiled. The hour of asking is a function inclusive of the hour required to access the surrounding fragment of data and the wiki view. This method is based on the concept that the work is carried out once, and the results are then analyzed again and again. The world's rich bases contain a completely new technology. The development of MBD contains only a few, as well as most new technologies. And they themselves are not as stable as relational databases (RDBs), and they themselves are not optimized. Another weakness of the MDB lies in the inability to analyze large numbers of databases during the data aggregation process, which requires an hour for new information to become available for analysis.

ROLAP.

ROLAP - tse Relational On-Line Analytical Processing, This is Relational OLAP.The term ROLAP means that the OLAP server is based on a relational database. The output data is entered into a relational database, either according to the “star” or “snowflake” scheme, which means faster processing time. The server will provide a rich data model for additional optimized SQL queries.

There is little reason to choose a very relational rather than a rich data base. RDB is a well-developed technology that has no scope for optimization. Vikoristannya in real minds resulted in more testing of the product. In addition, RDBs support more data obligations than MDBs. The stinks are designed for such duties. The main argument against RDB is the complexity of queries, the need to extract information from a large database using SQL. If there are any shortcomings, the SQL programmer can easily waste valuable system resources by trying to log in any similar query, which MDB can be logged much more simply.

Aggregated/Forward aggregated data.

The implementation of queries is imperative for OLAP. One of the basic principles of OLAP is that the ability to intuitively manipulate data leads to a rapid development of information. In general, the more calculations you need to make in order to extract a piece of information, the more likely you are to use it. Therefore, in order to save a short time for the implementation of queries, fragments of information, such as animals, are retrieved most often, and which require calculation, are subject to forward aggregation. Then they are protected and then saved in the database as new data. As an example of the type of data that can be obtained in advance, you can enter data - for example, sales figures for months, quarters or years, for which data are entered efficiently and are useful figures.

Different postal managers use different methods for selecting parameters to obtain forward aggregation and numbers of previously calculated values. The approach to aggregation is poured simultaneously into the database and at the hour of query implementation. As more quantities are calculated, the likelihood of a buyer requesting an already calculated quantity increases, and therefore the hour will be shortened, so that he will not be able to obtain a cob quantity for calculation. However, if we calculate all possible values - but this is not the best solution - the size of the data base naturally grows in such a situation, which needs to be compiled uncapped, and the time of aggregation will be large. In addition, if numerical values are added to the database, or if they change, the information must be displayed on previously calculated values that lie behind the new data. Thus, updating the database can also take a lot of time for a large number of previously calculated quantities. The data base operates autonomously during the aggregation hour, so that the aggregation time is taken care of.

OLAP -The client handles things differently. Pobudova rich world cube OLAP -Accounts are saved in the memory of the client computer.OLAP -clients are also divided into ROLAP and MOLAP.And actions can support the options for access to data.

Each of these approaches has its own “pros” and “cons”. Despite the broadening of the Duma's thinking about the importance of server functions over client ones, in a number of cases there is stagnation OLAP - Clients for clients may be found to be effective and profitable OLAP servers.

The development of analytical add-ons using client OLAP tools is a simple process that does not require special training. A developer who knows the physical implementation of a database can develop an analytical program independently, without hiring an IT specialist.

When installing an OLAP server, it is necessary to use two different systems, one for the creation of cubes on the server, and one for the development of client programs.

The OLAP client provides a single visual interface for describing cubes and setting up client interfaces to them.

So, in what situations might the installation of an OLAP client for business users be more effective and beneficial for the use of an OLAP server?

· Economical efficiency of stagnation OLAP - the server is at fault when the data obligations are too great and too much for them to handle OLAP -Client, otherwise it would be more justifiable to stagnate the rest. In this case OLAP -The client experiences high productivity characteristics and low productivity.

· Tightening the analysts’ PCs is another argument for self-righteousness OLAP -Clients. When frozen OLAP -Servers and efforts are not subject to vikorism.

The main advantages of OLAP clients can be called as follows:

· Vitrati vikoristannya ta suprovid OLAP -Customer cost is lower, lower costs are spent on OLAP server.

· When vikoristanna OLAP - For a client with a purchased machine, data transfer is carried out once. Under Vikonanne OLAP -operations will not generate new data streams

5. Principles of work OLAP-Clients.

Let's take a look at the process of creating OLAP using an additional client tool (Fig. 1).

Malyunok 1.Creation of OLAP-zastosunka using the additional client ROLAP-way

The operating principle of ROLAP clients is a forward description of the semantic sphere, which requires the physical structure of the output data. Some types of data may include: local tables, RDBMS. The list of data that is supported is indicated by a specific software product. Therefore, users can independently manipulate large objects in terms of the domain for creating cubes and analytical interfaces.

The operating principle of the OLAP server client is different. In the OLAP server, during the creation of cubes, the client manipulates the physical descriptions of the database. Thus, in the cube itself, inventories of the koristuvach are created. The OLAP server client is configured per cubic meter.

With the creation of a semantic sphere, the data sets – the Sales and Deal tables – are described in sensible terms and are transformed into “Products” and “Pleasure”. The "ID" field in the "Products" table is renamed to "Code", and "Name" to "Product", etc.

Then the “Sales” business object is created. A business object is a flat table from which a rich cube is formed. When a business object is created, the “Products” and “Products” tables are combined using the “Code” field for the product. To display the information, you do not need all the fields of the table - business object except the fields “Product”, “Date” and “Amount”.

Our application, based on the “Sales” business object, is based on the sale of goods over the months.

During the hour of work with the interactive sound, the user can set the filtering mind and grouping with the same simple “target” controls. At this point, the ROLAP client starts up with the data in the cache. The OLAP server client generates a new query for a large database. For example, if you are stuck with the product sales report, you can cancel the product sales report, which will help us.

All settings of OLAP programs can be saved in the available metadata repository, in an additional system repository of a rich database.The implementation depends on the specific software product.

Everything that comes with these add-ons includes a standard look at the interface, then the definition of functions and structure, as well as solutions for more or less standard situations. For example, popular financial packages. Further, financial programs have been created to allow specialists to access basic financial instruments without the need to design a database structure or behind-the-scenes forms and reports.

The Internet is a new form of client. In addition, he carried a lot of new technologies; impersonal Internet solution They really care about their capabilities and how the OLAP solution works. There is a clear advantage in generating OLAP data via the Internet. The greatest advantage is the need for specialized software to access information. This will save the business a lot of time and pennies.

6. Select the architecture of the OLAP program.

When implementing an information and analytical system, it is important not to compromise on the choice of architecture for the OLAP add-on. The literal translation of the term On-Line Analytical Process - “operational analytical processing” - is often taken literally from the sense that the data is quickly analyzed. This is a caveat - the efficiency of the analysis is not related to the actual time of updating the data in the system. This characteristic refers to the reaction time of the OLAP system to the customer's requests. In this case, data is often analyzed with a large amount of information “as of yesterday”, as, for example, the data in the collections is updated once a day.

In this context, the precise translation of OLAP is “interactive analytical processing”. The very possibility of analyzing data in an interactive mode differentiates OLAP systems from systems for preparing regulatory reports.

Another feature of interactive processing in the formulation of the founder of OLAP E. Koda is the ability to “combine, review and analyze data from the point of view of multiple variations, so that it is most useful for corporate analysts and in a way." For Codd himself, the term OLAP means, inclusively, a specific way of presenting data on a conceptual level - a rich one. While data can physically be stored in relational databases, OLAP tools typically work with large databases, in which the data is organized in the form of a hypercube (Fig. 1).

Malyunok 1. OLAP– cube (hypercube, metacube)

In this case, the relevance of this data is determined by the moment the hypercube is filled with new data.

Obviously, at the moment of forming a rich world-wide database of data, there is a lot of data that will be interested in it, so it is reasonable to delimit this obligation. Why not make it possible to analyze and not give the correspondent access to all the information that needs to be clicked? There are two alternative routes: Analyze then query and Query then analyze.

The followers of the first way are trying to gain access to a vast global database of updated information, for example, monthly, quarterly, and river supplies for children. And if there is a need to detail the data, it is recommended to formulate a report on a relational basis in order to place the necessary selection, for example, by days for which unit or by month of the employee m of the selected child.

The followers of the other way, however, will prove to be true, they will come across the data that they are going to analyze and capture them in a microcube - a small, rich database of data. The two approaches differ on the conceptual level and have their advantages and disadvantages.

Before moving on to another approach, it is necessary to introduce the “freshness” of the information, which the correspondent takes away from the world-famous name – the “microcube”. The microcube is formed based on new information from the current relational database. p align="justify"> Working with a microcube works in interactive mode - cutting out information and detailing within the microcube works instantly. Another positive point is that the design of the structure and the surface of the microcube is done “on the fly”, without the participation of the database administrator. However, the approach suffers from serious shortcomings. The koristuvach does not paint dirty pictures and is guilty of being directly involved in his investigation. Otherwise, the microcube may turn out to be too small and not contain all the data that needs to be collected, and your system will have to power a new microcube, then a new one, then another and another. The Query then analyze approach implements the BusinessObjects tools of the same company and the tools of the Company Outline platformIntersoft Lab.

When you approach Analyze then query, you collect data that is included in a large database of data, which can last for a long time, which may be subject to regulations and can last for a long time. However, all these shortcomings are paid off in a year, if the client has access to almost all the necessary data, no matter what combination. Reversing up-to-date data in a relational database is only possible as a last resort when detailed information is required, for example, for a specific invoice.

On the basis of a single rich data base, it is practically impossible to indicate the number of investors who work before it. You don’t have to read the data available there on the basis of the Query then analyze approach, in which the number of microcubes at the limit can grow out of the same fluidity, as well as the number of koristuvachs.

With this approach, there is an increased emphasis on IT services that, in addition to relational ones, are in need of servicing the world's richest databases.These services themselves stand for timely and automatic updating of data from a wide variety of databases.

The best examples of the “Analyze then query” approach are the PowerPlay and Impromptu tools from Cognos.

The choice of approach, and the tool that implements it, must lie ahead of the re-examined note: always have to balance between budget savings and the increased cost of servicing end-users. In this case, it is necessary to assume that, according to the strategic plan, the creation of information and analytical systems may achieve competitive advantage, and not the unique cost of automation. For example, a corporate information and analytical system can provide necessary, timely and reliable information about the company, the publication of which for potential investors will ensure insight and transfer These are companies that will inevitably become a source of investment profitability.

7. Areas of development of OLAP technologies.

OLAP is based on the task of analyzing multifactorial data. In general, given any table with data that needs one descriptive column (dimensions) and one column with numbers (entries and facts), an OLAP tool will usually be an effective way to analyze and generate data.

Let's take a look at the developments in the sphere of OLAP technologies, taken from real life.

1. Sales.

Based on the analysis of the sales structure, the necessary information for making management decisions is determined: about changing the range of products, prices, closing and opening of stores, branches, terminating and signing agreements with dealers, more advertising campaigns.

2. Purchased.

Pre-production analysis of sale. Many enterprises purchase components and materials from suppliers. Trade enterprises buy goods for resale. Possible tasks during the analysis were purchased without any responsibility, in the form of planning costs based on past evidence, up to control over managers, what do they choose postmasters?

3. Prices.

The analysis of purchases is followed by the analysis of market prices. The method of this analysis is the optimization of expenditures, the selection of the most profitable propositions.

4. Marketing.

During marketing analysis, we will pay attention to the area of analysis of buyers and clients and service providers. The objectives of this analysis include correct positioning of the product, identification of buyer groups for targeted advertising, and optimization of the assortment. The task of OLAP at the same time is a data-driven tool, because of the fluidity of thoughts, the removal of power supply, which intuitively arises in the process of data analysis.

5. Warehouse.

Analysis of the structure of surplus in warehouses by types of goods, warehouses, analysis of the terms of saving goods, analysis of the acquisition of goods and many other types of analysis that are important for the enterprise visibility in the organization of the warehouse environment.

6. Rukh koshtiv.

There is an entire area of analysis that is subject to an impersonal school of methods. OLAP technology can be a tool for the implementation or development of these techniques, rather than their replacement. We analyze the penny turnover of unprepared and ready-made currencies across business operations, counterparties, currencies and hours using the method of optimizing flows, ensuring liquidity, etc. The warehouse of vimirs lies in the peculiarities of business, knowledge, and methodology.

7. Budget.

8. Accounting documents.

The classic balance sheet, which consists of the number of accounts and the location of input surpluses, turnover and output surpluses can be better analyzed in an OLAP system. In addition, the OLAP system can automatically and quickly calculate consolidated balances of a rich organization, balances for a month, quarter and river, aggregated balances from a hierarchy of accounts, analytical balances for substitute analytical symbols.

9. Financial information.

The technologically advanced system of sound quality is nothing more than a set of named indicators with date values that need to be grouped and subsumed in different sections to isolate specific sounds. If so, the representation of other sounds is most simple and cheap to implement in OLAP systems. In any case, the internal communication system of the enterprise is not so conservative and can be overworked by saving money on technical work due to the creation of knowledge and the reduction of the possibilities of a rich operational anal Izu.

10. Promotion to the site.

The log file of the Internet server is rich in nature, which means it is suitable for OLAP analysis. Facts: the number of airdrops, the number of hits, hours spent on the site and other information that is available.

11. Obsyagi virobnitstva.

Here is another example of statistical analysis. In this way, you can analyze the quantities of grown potatoes, smelted steel, and manufactured goods.

12. Accumulation of expendable materials.

Discover the plant, which consists of dozens of workshops in which cold water, washing liquids, oils, ganchers, emery paper are used - hundreds of items of wasted materials. For accurate planning and optimization of expenditures, a thorough analysis of the actual consumption of expenditure materials is required.

13. Vikoristannya placement.

Another type of statistical analysis. Applications: analysis of the importance of primary auditoriums, which are available for rent, including premises, rooms for conferences, etc.

14. Length of frames for enterprise.

Analysis of the diversity of personnel for employment between phyla, branches, professions, level of education, status, century, hour.

15. Passenger transportation.

Analysis of the number of tickets and sums sold across seasons, routes, types of cars (classes), types of trains (flights).

This list is not delimited by the spheres of stagnation OLAP - technology. For example, let's look at technology OLAP - analysis of the sales sphere.

8. Vikoristan butt OLAP - technology for analysis in the sales area

Projecting a wide variety of data presentations for OLAP -The analysis begins with the formation of the vimiryuvan map. For example, when analyzing sales, you can completely see outside parts of the market (developing, stable, large and diverse partners, the likelihood of new partners appearing, etc.) and evaluate the sales volumes for products, Markets, buyers, market segments, channels and sizes will be established. They directly create the coordinate grid of a richly global representation of sales - the structure of its worlds.

Most of the activity of any business takes place in an hour, first of all, it comes down to an hour of analysis, and information about the dynamics of business development. Proper organization of the time axis allows for clear communication on the food chain. Call the whole hour to be divided into periods, quarters and months. Even more crushing is possible in the coming days. The structure of the time-hour data is formed according to the frequency of data acquisition; You can also consider the periodicity of receiving information.

The variety of “product groups” is divided in such a way as to maximally diversify the structure of the products that are sold. In this case, it is important to achieve a perfect balance, so that, on the one hand, you can see the exquisite details (a large number of groups can be accessible for inspection), and on the other hand, do not miss the relevant segment of the market.

The “Clients” category reflects the sales structure based on territorial-geographical boundaries. Each region may have its own hierarchies, for example, this region may have the following structure: Regions – Regions – Places – Clients.

To analyze the effectiveness of the activity of the products, create a moisture test. For example, you can see two equal hierarchies: departments and divisions that come before them, which can be seen in the “Parts” vimir.

In fact, the titles “Hour”, “Products”, “Zamovniki” continue to define the scope of the subject area.

Additionally, it is useful to divide this space into mental areas, taking as a basis the characteristics that are calculated, for example, ranges to suit the vartic expression. Then all business can be divided into low ranges in which it operates. In this case, you can use the following indicators: the amount of sales of goods, the number of goods sold, the amount of income, the number of requests, the number of clients, the total number of purchases from manufacturers.

OLAP – cube for analysis matime viglyad (Fig. 2):

Figure 2.OLAP– cube for analysis and sale

In OLAP terms, such a tridimensional array itself is called a cube. In fact, according to conventional mathematics, such an array will not always be a cube: a reference cube will have the same number of elements in all worlds, but OLAP cubes will not have such a boundary. The OLAP cube is not necessarily trivial. They can be both double and rich - depending on the task at hand. Serious OLAP products are available for a duration of about 20 years. Simple desktop programs support up to 6 times.

Not all elements of the cube have been filled in: since there is no information about sales of Product 2 to Client 3 in the third quarter, the value for the corresponding period will simply not be calculated.

However, the cube itself is not suitable for analysis. If it is still possible to adequately identify and represent a tridimensional cube, then from six or nineteen-year-old on the right means girsha. Therefore, before entering from the multi-world cube, the original two-world tables are drawn. This operation is called “cutting” the cube. The analyst takes and “cuts” the cube according to the marks of what to dig. In this way, the analyst selects a two-dimensional section of the cube (sound) and works with it. The structure of the star is presented by Malunku 3.

Figure 3.The structure of the analytical report

Let's slice up our OLAP cube and take a look at the sales data for the third quarter, which looks like this (Fig. 4).

Figure 4.Report on sales for the third quarter

You can cut the cube along each other axis and extract the sales data for product group 2 along the line (Fig. 5).

Malyunok 5.Quarterly announcement about sales of product 2

Similarly, you can analyze the data with client 4, cutting the cube behind the Client's mark(Fig. 6)

Figure 6.Notification about delivery of goods to client 4

You can detail the report by month or talk about the delivery of goods to the client's name.

Categories
- Android
- Program
- iOS
- Printers
- Editors
- Play
- Smart TV
- Internet
- Gadgets
- Zalizo

Popular