Experts advice : “It is pointless to split the nut with a sledgehammer.”-The same can be said about the data centers. It is better to match the chosen tool for the task. Typically the construction of such complex engineering of data center is based on the principle of entirety. A uniform model of the infrastructure used for the operation of all systems and applications, regardless of their censorious or production risk in the event of unplanned downtime. To meet the growing demand for critical operational environments using models that have a high level of redundancy. Data centers are designed to provide a compact technological systems, “uninterrupted” applications, and active business models. Nevertheless, in many cases this leads to unnecessary costs for the initial construction and ongoing operation of the data center.

While IT directors are forced to work within tight budgetary constraints, they should focus on next-generation data center. This center should include several working levels, each of which corresponds to the production priorities and the captiousness of specific technological systems and applications. Typically, multilevel data center reduces capital costs by 15 – 25%. This architecture saves energy, requires less support resources and work more efficiently, whereas it also reduce the total cost of ownership.

Development Of Data Centers

The data centers are developed in four-level system for measuring levels of reliability of different tier data centers. At the first level, data centers which have the lowest efficiency of 99,671%, while the fourth level of tier 4 are the data centers that have the highest rate of continued use of 99.995%. The dilemma, which has to be faced by the developers and administrators, is that, the most data centers are a mixture of technology and environment, which consists of systems and applications that have different levels of severity and business activity.

For example, bank ATM-systems must work around the clock, and therefore they need the fourth level, which consists of several power supply systems and air conditioning, redundant components and modules that provide fault tolerance. On the other hand, in the same environment, there are some applications, such as the verification system of checks and remittances, which are not heavily affected by the downtime and are suitable with a lower level of redundancy.

However, the standard data center have a monolithic structure in which all technological systems and applications are located in one place within a one-tier environment. The most critical processes that require a high level of redundancy, are scattered throughout the data center. One-level structures have no simple way to separate smooth and less critical applications. As a result, the entire center should always work “at full capacity,” with unnecessary redundant units reduce efficiency and increase costs.

Multilevel Hybrid Model Of Reliable Data Center

The next stage of development of data centers is a multilevel hybrid model, which allows a single data center to contain several working environments. According to this methodology, data centers can contain levels of a variety of opportunities that meet the specific technological and manufacturing requirements. For example, this model can be applied in developing the data center, the possibility of combining levels of tier 2 and tier 4. This allows you to select the appropriate level of hardware resources to a special section. All of these sections will be built in accordance with (technical) requirements of the specific levels of technological systems and applications that need to be placed in them.

These levels are determined by the method of structured analysis of production priorities.,for example, critical applications will run only in the sections that are intended for working systems that relate to the fourth level of continuity.

Proportional backup infrastructure, all of these sections provides substantial savings on capital (CAPEX) and operational expenditure (OPEX). The initial construction cost of data center can be reduced by 15 – 25%. Multilevel Data Center also provides increased flexibility and expandability. The modular “blocks” in the systems of power and cooling, data centers can quickly move in accordance with the requirements for redundancy in the different sections. This same system reduces the risks to business continuity as well as system and hardware failures are limited to smaller areas in the data center.

Conclusion

From the above explanation we can say that, the multilevel models meet the needs of the modern economy, allowing companies to operate within tight budget constraints, and also enable them to meet the growing production and maintenance requirements for many years.