DBMS: as TSO is estimated

Today many IT solutions are projected only within a single platform based on which all necessary complex of application and office software is created afterwards. Separate products are too expensive, both in terms of long operation, and in respect of upgrade. Something similar happens also to the database management systems (DMS) which are used for information analysis, formed owing to work of the modern enterprise.

In any large company terabytes of important business data collect. Therefore for their analysis and permanent replenishment DBMS which becomes a basis for long-term forecasting and daily orders over time is implemented.

However very often implementation of a new information system leads to the fact that the company stops the development and active market promotion. Most it is connected with the fact that in assets of the enterprise there is a minimum of available funds which are not enough even for new, obviously successful projects in the commercial plan. The reason is simple: the huge part of profit "is eaten" by overheads, daily expenditure for ensuring work of business processes. Eventually this financial article becomes more and more burdensome, however for radical upgrade it is necessary to completely change IT structure of the company that is often exigeant as requires the serious financial investments and often causes problems of technical character.

It should be noted that the most part of difficulties arises because at a stage of planning it was selected non-optimal for the available business processes of DBMS. The most frequent error – selection of the solution on the basis of only cost of the software license.

Unfortunately, the companies often forget that any DBMS is a complex multifunction product which is not possible for estimating within one criterion. Besides, the choice of a system which functioning will influence not only daily business processes, but also decision making from top management of the company it is impossible to perform, being guided only by the project cost or speed of work of DBMS.

What is included into TSO?

Let's consider a total cost of ownership (TCO, Total Cost of Ownership) DBMS, that boleechto finally it is possible to count efficiency of any information system, proceeding from the financial efficiency which gives its implementation.

Direct costs at implementation of DBMS are most noticeable - it is cost program and the hardware, the salary of the new personnel servicing its work.

Count implicit costs much more difficult as they directly are not connected with the IT sector in any way. Here it is necessary to carry accounts for an additional expense of the electric power, acquisition of the equipment for prevention of risks (restoration systems and duplications of data, the additional fire extinguishing system, uninterruptible power supply units, etc.), training costs of personnel of all levels (from operators and system administrators to analysts and top management of the company), losses from unplanned failures in an information system (violation of work of software, problem in the field of security, incorrect actions of personnel, etc.).

At the same time there is a set of the indirect moments which "cost" can be estimated only indirectly. To it refer efficiency of service of DBMS in comparison with other implementations of a system of collecting, the analysis and information storage, terms of an exit of cybersecurity corrections (payment of extra work of IT department), cost of development of separate modules for this platform, expenses on scaling of a system for the purpose of performance improvement, (need to buy additional software or the joint-stock company) and so on.

Thus, TSO include cost estimation, DBMS, necessary on all phases of lifecycle, for a long time.

What defines TSO?

Nevertheless it is worth emphasizing that IT departments do not carry out integrated value assessment of DBMS within TSO. Some pay the greatest attention to direct costs as a certain budget is one-time selected for it. As a result the one who implements the project has a desire to save (the manager often aims to cut down in any manner expenses).

Business became very dynamic, and only the few companies have faithful representation of what will occur in 3-4 years. As a result it is impossible to outline accurately perspectives of the company in the market and, respectively, to define what DBMS components will be necessary for it in the near future. The situation is aggravated with the fact that in many organizations communication between the managers (who are responsible for problem definition) and the technical department (realizing orders of the administration) is broken. Often there is no practice of structuring any IT problems and their data to the analysis of a normal economic situation.

As a result further work of DBMS is considered already absolutely separately, and all indirect costs (electricity, cleaning of premises, additional personnel, etc.) "are spread" in the general budget. Most often in the companies there is no practice of the differentiated calculation of similar expenditure from a point of zreniyakonkretny division.

Thus, the described problems connected with implementation of DBMS need to be liquidated for blueprint stage of the IT system. Otherwise there will be no economy in the long term. Therefore it is necessary to carry out integrated efficiency evaluation of implementation, carefully counting TSO.

It must be kept in mind that a segodnyana the market there are about four tens DBMS, each of which has the pluses and minuses which should be considered at implementation and calculation of TSO. Let's stop on four products - Microsoft SQL Server Enterprise, MySQL, Oracle Database and IBM DB2.

IBM and Oracle are the only companies today in the market which suggest to use not only specialized software, but also a hardware platform of own architecture (including corporate processors). It should be noted that such approach actually deprives of the user of a possibility of migration on other hardware platforms. It is possible for this reason need of expansion of functionality (at the expense of add-on software modules both from the DBMS developer, and from the third-party companies) is connected with significant financial expenditure and is not included in a basic set.

It should be noted that solutions for which deployment the universal hardware platform of Intel/AMD is used have advantage therefore the developers from Microsoft and MySQL who were initially relying on it receive much wider scope already in basic delivery of DBMS. However MySQL, despite the lack of a payment for the license, provides to users of a system a set of the difficulties arising at increase in volume of the processed information (for example, at data transfer on a cloud platform or in need of "hot" reservation). Therefore business projects in which deployment of additional functionality should happen quickly and without additional costs (both material, and temporary) should refuse use of such DBMS.

Thus, solutions in the field of data management which are proposed by Microsoft have much smaller TSO not only due to the functionality entering a basic packet but also thanks to new modules and the operational solution of non-staff situations (the questions connected with security, failures in work, etc.). Existence of technical support and the big knowledge base allows to achieve the planned goals with the minimum costs.

One more extremely important point is the fact that the Microsoft SQL Server provides communication with many databases (OLEDB, ODBC, JDBC, LU 6.2, XA ProtocolMessage Queuing, WebSphere MQ) and high interoperability thanks to transparent data exchange with products of such firms as Apple, BEA Systems, IBM, JBoss, MySQL, Oracle, SAP, Siebel, Sun and WebLogic.

In general, selecting DBMS, it is worth paying much attention to assessment of TSO. Otherwise the improper product can affect business of the company, and the question of transition to more optimal DBMS will be for firm more expensive the enterprise, than its initial implementation. To it it is necessary not only to purchase software, but also to retrain users (that it is sometimes much more difficult, than training from scratch).

In more detail on CNews