Where is the data center. The most complete information is that such a data center in the processing of personal data and its main functions

Data processing center - This is a building or part of which the primary function of which is to accommodate equipment processing and storage equipment, as well as auxiliary (engineering) tools that ensure its work (definition given in the US EIA / TIA-942 standard).

In the data center, powerful servers carrying out storage and processing of information are concentrated on a relatively small area; Network equipment responsible for data exchange with foreign world; Engineering systems that provide the vital activity of this "cybermaking" and security systems that protect the data center from unwanted invasions.

Fig. Data CCTP: Schematic diagram

Life support systems: ventilation, air conditioning, fire extinguishing systems, access control and video surveillance, structured cable system.

Servers and network equipment: Resource servers, application servers, information presentation servers.

INFRAMATION SECURITY SYSTEM: Anti-virus protection, spam filter, intrusion protection.

What companies need a data center

For a growing business sooner or later become characteristic:

• considerable increase in information
• growth of the amount of used business applications
• processing data in remote divisions from each other.

It comes time to consolidate data processing and centrally manage IT infrastructure and information systems - for this you need to build a data processing center (CDA).

TDACs are needed to all companies for which the maximum degree of readiness, fault tolerance, reliability of information systems are critical. These are large companies operating complex business applications (ERP-, CRM systems and others), communication service operators, banks serving client accounts and conducting plastic cards, insurance companies other.

Fig. What companies need a data center

Now in Russia the number of data creation projects is growing, the codes are complicated and increasing in size. The advantages of using data centers begins to realize the average business. Long acquired by them, large companies are increasingly resorting to the creation of reserve capacity. In general, the system integration market is now evolving in the direction of creating networks of data centers.

CDA requirements by industry

The traditional consumers of IT services were and remaining enterprises where information technologies are critical for business, and the execution of business functions directly depends on the level, quality and degree of availability of IT services. These traditional consumers included state structures, banks and telecommunication companies. Today it is mature, from the point of view of the development of IT, users with the current culture, approaches and understanding of the place information technologies In the company hierarchy or organization. From a technical point of view, these are formed codes equipped with modern equipment and software. They are characterized by another spectrum of problems: where to take electricity, as in the context of the widespread lack of qualified IT-personnel to build an effective service service that can serve the maximum large equipment park with a minimum number of employees, how to keep specialists, etc. For such companies and organizations are important and Other issues related directly to their business: for example, how to ensure the integrity of the IT structure when buying one enterprise to others? Or how to divide the holistic IT infrastructure when dividing the organization?

Today in Russia there is a new round of the development of information technology, when industrial enterprises, retail become the most "greedy" trading networks, Insurance companies. It is here that the greatest interest in IT is observed, in particular to the Codes. Companies are in the permanent search stage: where it is better to place your data center, which software tools will be able to solve the tasks that the optimal hardware platform for the operation of the necessary applications.

Thus, the criticality of information technology for business is aware of the players all more Market segments, IT penetrate into the economy of enterprises, increases their role as a business tool. In other words, the market today is on the rise, and companies - system integrators are enough.

A source: IT Manager.

Advantages of the introduction of database

Unlike a decentralized approach to the organization's IT infrastructure organization, the availability of data saves finances and increases:

• reliability all information system (reliability of data storage, equipment fault tolerance and software)
• the level of services provided by the Company to its customers
• the labor productivity of employees by increasing the speed of operations, improve control, etc.

In addition, the data center provides opportunities:

• modernize and build computing systems in the conditions of introducing new business applications
• centrally manage IT infrastructure and information systems
• reduce the cost of ownership of information systems.

Corporate and hosting data center

The corporate data center is initially created to solve the tasks of the automation of business processes of the customer himself and the owner of the CDA.

The hosting data center is leased: the owner of the Data Center allocates stall-place organizations or clusters that are filled with tenant equipment.

The mixed Code is partially focused on ensuring the owner's business processes, and partly to solve the tasks of tenants.

Main and Reserve Code

The main coder of the information and telecommunications system. He assumes all the load in normal mode.

The backup data center ensures the usual provision of services in case of failure, prevention or hot replacement of equipment installed in the main data center.

Place Code in the company's IT infrastructure

The data center is a strategically important node of the organization information system.

Code provides:

• consolidated processing and data storage
• maintain a specified automation mode of business tasks
• the preservation of corporate information is usually high commercial value.

From stability, reliability, timeliness, efficiency and completeness of its services directly depends the success of the activities and competitiveness of the customer.

According to META Group, financial costs in the event of refusal of systems and equipment are:

• $ 6.5 million per hour in case of refusal of systems providing brokerage operations
• $ 2.6 million per hour in case of refusal of credit card authorization systems
• $ 14,500 per hour in the event of an ATM failure
• $ 330,000 on average cost per hour of downtime data center.

Announcements

Code is decrypted as a data center. Many still do not know what it is. Meanwhile, he is whole complexDesigned for storing servers and equipment and ensures their uninterrupted operation. Most often, such centers are located in large buildings and include large-scale servers.

Now the services of the data center are also popular among young companies whose investors are foreigners. In Russia and the CIS countries, the popularity of such centers is still low, but they are considered profitable abroad.

Difference from the database

Practically in any company there is a database to optimize the workflow.

This base can be stored in different ways, and access to it must be uninterrupted. However, various factors, such as temporary absence of electricity, or software failure, can affect the data processing speed, or temporarily restrict access to information.

Reference. The processing center provides conditions in which the equipment will work continuously, without such failures, in comfortable thermal conditions. This is the main difference between the CDA from the server room.

Views

Data processing centers are two types.

1. Local. In this case, the room with a group of servers is in private property and under private control. The owner of the server independently places the data center in certain premises and its independently serves it.
2. Commercial. Such centers are large-scale projects aimed at leasing equipment to large corporate customers. Most often, the services of commercial species use companies that for one or another can not ensure uninterrupted operation of the servers. It is more profitable for them to rent a center for processing, providing information to reliable storage and proper protection.

Goal

In addition to the continuous operation of the equipment, the data processing center must allow:

• increase the reliability of information storage;
• ensure all applications on one equipment;
• have sufficient capacity, the necessary types of equipment;
• reduce the cost of service IT infrastructure.

Commercial data center, unlike local, exclude the incorrect organization of the system, and also make it possible to save on the content of the staff of specialists. In addition, there is no need to highlight one or more premises for the operation of data centers.

What is the Date Center when processing personal information?

Server

The operation of the data processing center provides high quality high speed information At the same time, without losing the details and keeping the integrity of this information. Server performance is one of the main factors that the company pays attention to.

In addition, the important features of the century as the server are considered:

• the ability to cope with large loads in real time;
• high degree of manageability;
• permanent access;
• redistribution of loads in order to more quickly solve business tasks.

As service

Companies providing services of data centers offer a wide range of basic and additional services. They include leased telecommunication racks, cabinets, selected servers, colleysis, virtual hosting and the ability to use cloud platforms.

Important! The center must have external security, ensure access control and provide a guarantee of reliability.

Moreover, an important factor in the work of the Data Center is subject to sustainable Internet channels. and the presence of traffic exchange points. Power supply should be no less reliable, include independent electricity inputs and diesel power plants.

Thanks to the above factors, the data center as a service can be considered highly efficient and reliable, having a fault-tolerant infrastructure.

What is notification about the location of the location and how to specify the address?

The company must send In which you need to specify information about the location of the database of information containing citizens Russian Federation (According to a change in 152-F3 of 21.07.2014 N 242-F3).

You can send the document to the structure both on paper and in in electronic format. Mandatory signature of an authorized person. The notification must contain the following information:

• name (FULL NAME) and its address;
• data categories;
• subject categories whose data is processed;
• list of actions with data, processing methods;
• information about the availability of encryption funds, their names;
• FIO of the individual or the name of the legal entity, their contact details, postal addresses, email;
• data starting date;
• deadlock or conditions of processing;
• information on transboundary transmission (its presence or absence) in the processing process;
• information about the location of the information database.

Filling in the last paragraph should contain information about the location of the DB of information with personal data of citizens of the Russian Federation. You must specify the country and the exact address of the central address, if it is local, or provide information about the owner of the center, if it is commercial. In the second case, you need to provide the following information:

1. type of organization (physical or entity, individual entrepreneur or foreign organization);
2. organizational and legal form;
3. name of company;
4. OGRN;
5. the country and address of the location of the organization.

Important! It is necessary to indicate the storage locations of personal data of citizens of the Russian Federation and only.

The rational application of the CDA will allow large companies to reduce the costs of time and means, Provide proper security storage and uninterrupted access to it. Choosing a local type of data center, it is better to make sure of the above objectives, and when choosing a commercial, it is important to view all licenses confirming the quality of the service.

Didn't find answering your question? Find out, how to solve exactly your problem - Call right now:

To date, no one is not surprised by the fact that in a particular company there is a private data center. What it is, only a few employees of this organization know, but in reality such equipment requires any business that wants to achieve real stability. In other words, if there is a real need to ensure uninterrupted, scalable and manageable work of the company, when the business stability of the business directly depends on the IT infrastructure, the data center is used directly.

What is it?

Thus, over the time and development of information technology, in almost any organization, which is somehow related to information, there is a private data center. What it is? The data center, which in professional specialized literature is often customary to call a data center. From the title, it can be understood that in such equipment a variety of operations are carried out, which directly relate to the processing of any information, that is, creating or generating data, subsequent archiving and storage of files, as well as subsequently providing them at the request of the user. Wherein special attention In addition to the above functions, there are also safe destruction of data for which the data center is also responsible. What it is? Eliminating certain files without harm to other data and perhaps without the possibility of recovery, if really important information is deleted, which should not get into third hands.

Where can they apply?

To date, there is quite a large number of funds, including land inventories, Pension Fund and a variety of libraries that are directly engaged in the collection and storage of various information. It should be noted that there is also such information that in itself generates a business, for example, the one that is applied with a variety of reference services. There is also information that does not take any participation in business processes, but it is necessary for their implementation. Such data include personnel service files, as well as user account databases in various information systems.

Industrial holdings are created by specialized electronic archives to perform settlement tasks, storing documents, as well as automation of business processes. Thus, different organizations use different types of information, as well as tasks that concern its processing. It is for solving such tasks and a data center is created. What it is, knows only system Administrator, on the shoulders of which the equipment lies.

When is the data center?

Tasks associated with in different time Distened using various technical means. In the twentieth century, electronic computing devices became the basis of modern business, since they assumed the most computational tasks, and the appearance of devices that store information provided the opportunity to completely get rid of paper archives, replacing them more compact, and at the same time available electronic and tape carriers. Already in order to place the first electronic computing machines, it was necessary to highlight the specialized machine rooms in which the required climatic conditions were preserved so that the equipment does not overheat during the work and at the same time it worked stable.

Server rooms and their features

With the beginning of the development era personal computers And small servers computing equipment of almost any company began to consist in special server rooms. In the predominant majority of cases under such a room, a certain room is envisaged in which household air conditioning is installed as well as source uninterrupted power In order to ensure the continuous operation of the technique in good condition. However, these days this option is suitable only for those enterprises in which business processes are very dependent on the information used and available computing resources.

What is the difference between the data center from the server?

By and large, the modern data center is an extended copy of the traditional server room, because in fact they have a lot of common - this is the use of engineering systems that support the continuous operation of the equipment, the need to provide the desired microclimate, as well as the corresponding level of security. But at the same time there are a number of differences that are decisive.

The data center is equipped with a complete set of various engineering systems, as well as specialized components providing normal and stable operation. information infrastructure Companies in the mode that is required for business.

Where is such equipment used?

In Russia, data processing with such centers has become in demand since 2000, when such equipment began to order various banking structures, government agencies, as well as enterprises in the field of oil industry. It is worth noting that for the first time the data center appeared in 1999, when it began to be used to process declarations and certificates of income of all residents of Moscow and the region.

Also one of the first large centners was the equipment that was used in the center of Sberbank. In 2003, with the support of Rostelecom, the first republican data processing center was organized in Chuvashia, used to systematize archival data. Such devices provided various local authorities, and in 2006, the Center also opened, in which the processing of the Data Curchatov Institute was handled. Next year, VTB-24 and Yandex also started using their own data center. Moscow, therefore, quite quickly came to the use of such equipment, as well as other major cities of Russia.

Where is it worth installing the data center?

Nowadays, its own dwarf is used by almost every large geographically distributed company, especially if the business is very dependent on the organization IT. As examples, you can bring telecom operators, company retailers, tourist and transport companies, medical institutions, industrial holdings and much more.

Code can be intended to work for a particular enterprise or to be used as multiplayer equipment. The multiplayer data center provides a wide range of services, including business continuity, as well as hosting, rental and server location and many other elements. The data center services are the most relevant for small and medium-sized businesses, since it can be eliminated by the need for modernization of IT infrastructure, and ultimately get a guarantee of reliability and the highest quality service.

The key to the success of the data center is competent design

Competent design of the data center allows you to eliminate the occurrence of serious problems in the process of operation of the equipment, as well as reduce costs during operation. In general, the structure of such a center is distributed to four main elements - this is an engineering infrastructure, a building, software and specialized equipment. At the same time, the construction of buildings and premises for the installation of such equipment is executed in a variety of standards, the main purpose of which is to ensure safety and reliability. In Western countries, extremely serious and sometimes very peculiar requirements for the data center are often established - in particular, it is worth highlighting what the building must be at least 90 meters away from the extremely high point, which reached water when flooding over the past 100 years, which It is not so easy to achieve as it may seem at first glance.

Maintains information on the already implemented data projects, or by advanced technologies used in its creation. For some reason, the question of the substantiation of the selection of a data center, questions of competent compilation of TK on it, as well as questions about effective use All the possibilities of the data center put in the data center turns out to be missed. I will try to highlight your forces and opportunities to highlight these questions.

Scope of the document and the list of issues

This document is designed to obtain a complex of necessary information for specialists in the creation and operation of the data center, server and machine rooms.

The document discusses:

• Problems that appear on the design stages, construction and operation of the data center, as well as possible solutions These problems
• Recommendations on the use of modern standards are given, as well as their brief characteristics.
• Basic errors in design, and problems appearing during operation are shown their consequences, as well as possible ways to eliminate errors and solving problems
• Separately, the rules for creating successful IT projects are given.
• The most important requirements for the main elements of the data center, and, if possible, explains the reason for these requirements and the consequences of their non-compliance.
• Listed key trends in creating a data center and some statistics of foreign and Russian data center

Of course, this is not a comprehensive document, and within the framework of one document to consider the main issues arising at the stages of justification, design, commissioning and operation itself is not possible. Therefore, I will try, if possible, refreshing the key points of the entire life cycle of the Data Code, to pay special attention to the issues, in my opinion, the least described in the literature and the Internet. From the fact that some questions did not receive proper lighting, they do not cease to be important, especially since some of them, as shown will be further silent for quite definite reasons.

I will try to clarify the circle of specialists for whom the document will be oriented. These will be specialists of organizations that do not have a data center, but those who want to build it, the specialists who have decided to build a data center, but not know what to pay attention to when writing a technical task (T) s and how to choose a partner, specialists who built the data center, but trying under the ego of operation Ensure the stated characteristics and reduce costs. And also the document will be interesting to equipment suppliers and data developers, at least in terms of understanding the problems of their clients. Although the document will also consider the majority of questions arising from the substantiation of the choice of data center, its design, construction and operation, but in the document there will be no instructions to choose one or another equipment, and even on the mandatory use of certain technologies. The fact is that new technique, solutions and technologies appear every year, often, in essence, differing from some non-existential changes, or the implementation of long-known solutions, but on the new technical level. Remember - " Knowledge of several principles, frees us from knowledge of a set of private " Based on this, I will first try to tell about the principles of design and operation of complex computing systems, which, as it is not better suitable for the data center.

In order to discuss the problems of building and operating the CDA, it is necessary to decide on some terms and understand what kind of data center. Therefore, at first I will try to decide on the term "CDA".

Definition of the term "CDA"

Recently, it became very fashionable to argue about creating a data center. Almost every self-respecting company declares that one of its specializations is the construction of a data center or data centers. Usually companies refer to positive reviews, completed projects, etc. etc.

Let us first try to figure out what kind of data center than it differs from just a good server, as well as what properties of the central system allow it to be called the Code. In the same way, we will try to understand, the execution of which works when building a data center requires special attention and where you can save without loss in quality. Analysis of all this will allow not only to create a better data center, but also it is useful when building other storage and data processing objects.

If you turn to Wikipediathat Dataor centre storage I. data processing (Data/Cycling) is a specialized building for placement (hosting) server and communication equipment and connecting subscribers to Internet channels. Another name of the data center - Data center (from English. data Center.).

Comment : If the document shows the term " Data center"This means that it is cited, or the document is retelling where it is similar to such a term, and not the term" Data».

In fact, such a interpretation at least does not reveal the whole of the essence, what is such a data center. Significantly closer in meaning Such a interpretation "Code is a building (or part of it) for which comprehensive storage solutions, processing and dissemination of information data with IT infrastructure allowing to provide their functions satisfying certain criteria"

In any case, in determining the data center, it is not necessary to emphasize the presence of hosting and Internet, because They can really be, but their absence is not critical for the data center. In the form of the specified formulation of the data center, it most fully corresponds to the concept of the data center set out in the standard TIA-942.. Although, in my opinion, should be clarified by the wording " Code is a building, its part, or group of buildingsFor which ... "Next by text. Because It may well be that in the implementation of a data center with duplication of subsystems, a geographically data center will be separated between several buildings. Sometimes they remember that when operating a data center, it is necessary to develop a complex of organizational procedures and constantly engage in personnel training. But this is no longer so principal, because It is only worth understanding that the Data Code is not only a building, but also the complex engineering solutionsYes, and not only it, as well as ensuring the necessary services and the availability of qualified personnel.

Historically, data centers (the name of the data center appeared later in Russia) rose from large server servers available from IT companies in the 90s. This qualitative change facilitated the emergence of client-server technology, the emergence of new cable network standards, and the appearance of hierarchical control of data carriers. The main features of the data center were formed by the year 2000, when the data center were very in demand for deploying Internet servers of organizations that do not have opportunities for their support, as well as ensuring work in their computing centers of the databases of various organizations.

Currently in one St. Petersburg more 30 Code. In fact, they are more, because Some organizations have built infrastructure suitable for the concept of the data center.

Regarding standard TIA-942. It should be noted that in the document in detail the issues of construction (mainly in the form of presentation of the requirements) of engineering subsystems, but if you try to ask the choice of a specific project to build a data center, in order to fulfill specific tasks, immediately appear questions. In the standard TIA-942, the concept is introduced tier levels. The standard considers four levels associated with varying degrees. ready (terminology TIA-942 ) Infrastructure equipment of the data center. Higher levels correspond not only to higher readiness, but accordingly cause increased costs of infrastructure. In fact, the TIA-942 standard shares (classifies) the codes only in terms of reliability (sometimes they write that in terms of availability, but this term is also close, but still he is already a term " reliability»).

Classification Center

The very concept of a data center is quite small-informative, the case is all the centries are different not only in size, but also on the tasks assigned to them, if possible, to provide their main functions with a certain level (quality). And the basic functions of various data modes, depending on their orientation, various functions may be considered.

If you look more carefully, you can select quite a lot of criteria for which you can split the centries. Basically, it is precisely these criteria that will be determining in the functioning of the CDMS, or these criteria will carry a set of some properties to allocate a specific group of the CDMS.

You can divide the codes by:

• Appointment or rather - divide them into public and not public (the term "corporate") centners are used;
• Reliability of data storage (if you become more accurate, then the totality of reliability and availability).

There are still as separate groups CatastrophorestableData processing centers (QCD) and " tRACH DATA CENTERS" The name "Tresh" went from (eng. tRASH - garbage) - usually these are small codes, cooling in which is implemented only at the expense of a natural air exchange.

Such "garbage" codes most of them do not meet the fully requirements for the sets, but less costly, environmental and rental of server racks are significantly cheaper.

With the division into public and non-public data center, everything is clear, and they have a different approach to design. After all, making the data center for themselves, the organization knows enough well which of the main properties is needed, and where it can save. Hence the possibility of selectively executing requirements for the data center. In the public data center, everything is somewhat more complicated and if they want to receive certification to increase the number of their customers, then, at a minimum, mandatory recommendations will have to perform everything.

If we talk about reliability, it is necessary to start with the consideration of the term "working on failure". Actually, it is not a fact that the system after the failure of the failure of one of its elements will cease to function. If, when exiting (transition from a working condition in not workable) one of the elements of the system, the system becomes inability, then they say what happened renouncement. If still the system remained working, then they say what happened failure. The moment and frequency of failures and failures are described by methods of probability theory and in this document is not considered. The only thing you need to remember that, only analyzing scheme The reliability of the system and having data on developments on the digital expression of each composite part can be said the level of availability or performance of the entire system. The share (%) of the time during the year, when the system is in working condition and / or in idle state (% Uptime and Down Time), directly related. Downtime period (Downtime) is the total indicator of downtime for the year. These terms often operate when discussing different levels (Tier.) Code. But their digital expression for different levels is not correctbecause The scatter of reflaking indicators from a data center of one level can be great. In the appropriate place of the document it will be shown that all the figures characterizing the period of downtime at various levels of the CDA from the evil and actually do not really rely on them. If we talk briefly, the list of the most characteristic features of various data levels can be reduced to a simple table.

For example, I bring the compliance of the availability, the time of finding the system in non-working condition (per year). I will not bind the levels to the numbers, because I have already said above, the scatter of availability in the year may be within one level great.

The appearance of the type "without reservation" does not mean that in the event of an exit of the establishment, an order and receipt of the refused block will be expected. The presence of calculated zip reserves and the decrease in the value of the MTTR indicator (average repairs time) also significantly affects downtime.

Another important remark. Code will be the maximum level of which minimum level of one of the components. But on the other hand, you need to remember that not all recommendations from standards are mandatory, and if you know exactly what and how the violation is affected, you can usually save somewhat When building a data center.

Example

Developers, often struggling for improving the energy efficiency of the data center, which,it is estimated as the ratio of total power to the power of IT equipment for a long time fought for the ability to increase the operating temperature. The idea of \u200b\u200bhello, because the life of most computer equipment in the center of 3-4 years, although it is generally necessary to note that the equipment is responsible for energy supply is usually replaced less frequently, however, with the right maintenance. After this period, either the equipment is replaced or the most critical applications are transferred to another new equipment. Increasing a few degrees of temperature in the room really does not affect the likelihood of equipment failure for this period, but significantly reduces cooling losses, thereby increasing energy efficiency.Now there are trends for some CDA classes to still increase the permissible temperature.

Therefore, it is very important to know about why are those or other requirements in standards, and what will be with a deviation from the standard in one direction or the other. With all this we can figure out, only analyzing the requirements for topics or other parts of the data center. It is also necessary to deal with the question of which standards regulate the requirements for constituent parts of the Data Center, whether they are not contrary to each other, and in general whether these standards should be observed. Therefore, the next chapter will be devoted to standards and their requirements.

Requirements of standards for components of the data center

Initially, it is necessary to determine the requirements of which standards it is necessary to be guided, and most importantly - what will happen if they are somewhat "broken", respectively, for the best or worse. At the very beginning of the chapter, I will express a few wise thought. Standards need to know in order to be able to violate if necessary if necessary. More accurately reasonable to do some of the requirements for your specific data center above or lower than the requirements of the standard to the class of the CDM class. I wrote this line and understood, now it will definitely have to write the name of this "smart" standard, the requirements of which it is necessary to follow when developing a data center. But ... no - not everything is so simple. Documents carrying in their title proud name "Standard ..." are actually most often the generalized experience of a group of experts created this standard. To accessibility (% Up Time) or idost time (DowN.Time.) recommendations are not directly related. Following the requirements of standards really allows you to improve these indicators, but for which magnitude, this is a mystery covered with darkness. The fact is that it is practically not possible to take into account all the factors affecting the reduction or an increase in these indicators and, the more impossible to obtain data on the whole used specifically the equipment of your data center. What to do? First of all, to put on the priorities, the requirements of the CDA you create to try to take on the basis of one of the standards and further follow its requirements with possible accuracy.

In my opinion, to start a search for a suitable standard for you need from previously mentioned TIA-942. « Telecommunication infrastructureData Centers ». The first version of the standard was published in 2005. It detailed details the requirements for constructings, energy supply, heat sink, security control, redundancy, maintenance and commissioning of operation.

In June 2010, Building Industry Consulting Service International Inc. (BICSI) published a new standard 002-2010 : Data Center Design and Implementation Best Practices. This standard BSCI 002-2010 Reflects the increase in the complexity of the design of computing centers and the need of companies and organizations in understanding the requirements for energy, mechanical loads and telecommunications in the design of the infrastructure of the MC.

What is the standard better to use? What are their differences? How then is certified? After all, there are standards from other organizations. For example, the main difference in certification according to Uptime Institute standards (Institute of Uninterrupted Processes) is that the certified specialists of this organization should make sure to implement the requirements set out in their standards. In mid-2010, Uptime Institute has released another standard " Operational Sustainability (Operating stability) "Regulatory and operation services. It is not enough requirements for the service of operation in TIA-942 . And although together fulfilling the requirements of the standard TIA-942 I.standard Operational Sustainability It is already enough to accurately formulate the requirements for the data center, but in practice, the builders of new computing centers are more likely to refer to the standard TIA-942. The fact is that each of the standards was compiled by a different organization and in many details differ from each other. Especially since, according to UPTIME Institute experts, their separation procedures are not functionally connected to the levels of TIA-942, they estimate the ability of computing centers to maintain performance in fault conditions and accidents. To avoid confusion, UPTIME Institute experts propose to designate the levels of accessibility in their interpretation of Roman numbers I, II, III and IV. It is quite difficult and certified by the data center. If you contact the site Uptime Institute. (site http://uptimeInstitute.com) then at the end of May 2012 really provides IV level (i.e. not only documentation and created building with technical means in it, but also the level of operation) only 1 center, certification of the constructed object on IV Tier was carried out for 6 codes. Certification of documentation for building a data center IV Tier is obtained for 22 objects. Russian data center among Tier IV on currently not. Todhov level III is also not very much. Provide full Performing requirements for III levels for "operating stability" of only 4 centies. Russian among them are not. Documentation and premises correspond to Tier III in 5 Russian data center (4- Design Documents and 1 Constructed Facility).

During 2012, the TIA-942-A standard will be published, which included changes and additions to the following versions of TIA-942-1 and TIA-942-2. Unfortunately, a new version The standard has changed strongly. The new standard TIA-942-A will concern only the topics of cable systems and will no longer be such a comprehensive, which was the standard TIA-942. Those. Basically ON. will regulate only the construction of cable systems. The energy efficiency section is likely to concern this topic only from the point of view of the cable system and the use of the "green" data transmission environment - fiber.

Below is a list of the main changes included in the current TIA-942-A project (according to the preliminary application of the developer). This information is in italics.

TIA-942-A is aligned with a series of TIA-568-C standards for topology, terminology and environmental classifications presented in Standard 568-C.0, as well as component specifications presented in TIA-568-C.2 and C .3;

• Applications, TIA-942-1 and TIA-942-2 are included in the TIA-942-A standard;
• Information on the ground was moved from the TIA-942-A standard to the TIA-607-B standard;
• Administration Information will be moved to the TIA-606-B standard;
• Most of the information relating to telecommunication cabinets and server racks, separation of power and telecommunication cable systems will be moved to the TIA-569-C standard;
• Information on the outer cable wiring is moved to TIA-758-B;
• The limitation of the length of horizontal fiber optic cable systems is canceled with 100 meters.
• Category 3 and Category 5e cables should no longer be used in horizontal cable systems. The working version of the standard is allowed to use balanced twisted par Category 6 and Category 6a in horizontal cable systems. Category 6 and Category 6a can be used in main cable systems;
• Application in horizontal and main cable systems of multimode fiber optic cables type OM3 and OM4 (multimode optical fiber with core diameter / shell 50/125 μm, optimized for operation with light-based light sources on the wavelength of 850 nm). OM1 and OM2 cables are no longer permitted for use;
• To connect one or two fiber cables, fiber optic connectors of the LC type and for mpo mole-fiber connectors should be used;
• The Topology Code includes an intermediate distribution area (IDA);
• The standard has added a section on energy efficiency;
• The terms "hardware socket" are added (EO - Equipment Outlet) and "External Network Interface" (ENI - External Network Interface) borrowed from the international standard ISO / IEC 24764.

STANDARD "OPERATIONAL SUSTAINABILITY (OPERATIONAL SUSTOST)" just complements TIA-942 Especially in the operation of the data center.

The "Operational Sustainability" standard describes the requirements to ensure the stability of data centers, as well as minimize risks associated with this. As you know, the preceding widespread standard "Tier Standard: Topology" regulated the technical parameters of the data center necessary to achieve a certain level of reliability. The feature of the new standard is that he gives the human factor In the sustainable work of the data center. And it is important, since the percentage of errors in the work related to this factor reaches 70% , of them a little more 40% related to errors managerial management services. To minimize these errors, it is necessary to conduct targeted work with personnel, raise its qualifications, conduct activities to hold qualified personnel.

If we consider standards from the corporation Bicsi., It can be noted that their approach differs from approaches to assessing the levels of sustainability of other organizations.

Sustainability level assessment system and basic standard sections BICSI 002 2010. . According to the Association, the Standard developers set themselves the goal to ensure the design and construction of data processing centers, taking into account the long-term perspective of their operation. The main sections of the document:

• Planning Data
• Select site
• Architectural solutions
• Building construction
• Electrical systems
• Mechanical systems
• Firefighting
• Safety
• Building automation systems
• Telecommunications
• Information Technology
• Commissioning
• Operation and maintenance
• Design process
• Reliability

Therefore, about standards for building a data center, it should be noted that all developers of general standards for the data center do not contradict each other in terms of requirements and references to standards when building basic data levels. Commercial data center due to their specifics should satisfy (and preferably be certified) all the requirements of the standard they have taken as a basis. Not all recommendations affect the basic quality of the data center - ensuring the specified availability level. Therefore, non-profit data center in a number of cases can some requirements and ignore. Especially since certification thing is not only expensive, but also really does not affect the level of efficiency of the data center. After the data center is implemented, you can still make some changes not only to the level of support, but also to other levels, trying to meet the requirements of some of the standards for obtaining certification.

UPTIME Institute at one time defined four levels related to different degree of readiness of the infrastructure of the equipment of the data center (CDA). In fact, even though they are associated with the level of availability, but probably more correctly talk about the levels of Tier, although the term "Tier" itself and translates as the "level". Above me, I do not in vain revealing the concept of "level", did not lead digital characteristics The availability of the data center. Digital expressions were obtained only from the analysis of implemented projects. I give some data from the document developed by the Institute of Health Problems (The Uptime Institute) in the Bulletin "classification by the industry standard, which determines the quality of the object's infrastructure work" (Industry Standard Tier Classifications Define Site Infrastructure Performance).

General conclusion on the use of standards:

• The fundamental should be considered the use of the TIA - 942 standard with the latest additions (for example, with the standard "Operational Sustainability (Operating Stability)";
• The new standard TIA-942-A (approved on April 24, 2012) concerns only the topics of cable systems and will not be such a comprehensive, which was the standard TIA-942;
• When building a data center, it is necessary to use not only standards, but also common sense, allowing to significantly save, not worsening the most popular qualities;
• Certification is more necessary for the commercial data center, and the center of the organization may not be engaged. Of course, if the data center is, it was created on the basis of standards, then all deviations from recommendations must be reasonable;
• To read, and, most importantly, to understand what standard to take as a basis and on what requirements it will be necessary to make an emphasis in future development, it is impossible to assume that you have completed work with the standards. Before moving to the next stage, it is necessary to reread old, good, really, at the moment, mainly forgotten GOSTS - series 34. And nothing that they have not been updated for many years, but there is a detailed consideration of pre-project stages. They do not have the words "business processes", "processor approach", but there is a notion " information model»It is quite correctly replacing them. Therefore, especially at the TK stage, these documents will help you. Of course, it is necessary to approach creatively and not follow the literally to all the recommendations, but carefully read them.

Procedure for constructing a data center

As not strange the greatest contribution to the success or not successfulness of the future project makes the initial stages. In fact, in world statistics in the IT industry it is successful only - one project from 3. If you get more rigid, and evaluate the success of the project as:

• the ability to execute the claimed functions with the required quality
• perform work for the planning time
• the nebid beyond the initial budget of the project
• lack of revolutionary work at various stages of the project
• no need to immediately begin work on the modernization of the project.

Everything will be worse. Probably, no more than 20% of projects will fall under the definition of "successful".

There are many reasons for failure of the project. There is an incorrect policy (it is the policy, because the decision of the controversial issues is most often found by compromises) project management, lack of due support from the head of the organization, weak developing TK and as a result of a large number of unplanned work, a weak participation of the organization's specialists for which the project Any force majeure is performed.

If almost over each project brings the probability of failure, how to deal with vigorous statements about dozens of successful projects in many firms? First you need to immediately put everything in your place, determining the term " Project».

Project (If you turn to Wikipedia) - this is a unique (in contrast to operations) activities that have the beginning and end of time in time aimed at achieving a predetermined result / goal, creating a certain, unique product or service, with specified restrictions on resources and timing, as well as quality requirements and permissible levels. Risk. Probably this definition can be simplified for greater specifics. Project— this is a totality of tasks, activities or work performed by the achievement of the planned goal, which usually has a unique and non-refining character . The main thing is that the project is always unique (at least for persons performing it). So everything is what performers say as a successful project is actually successful introductionthose. Implementation of a ready-made solution. The percentage of the implementation of successful implementation is significantly higher than successful projects. And if the programmers writing any complex program There is always a project, then introduction is possible in the field of building infrastructure. It is quite difficult to carry out the face when the implementation of the implementation will develop into the project. For example, if a small software and hardware complex is created for automation, some remote site and is done by the developer not for the first time, and the number of differences from previously created both in the hardware and the set of installed programs is minimal, then this is the introduction. And it has a fairly big chance of success. If there are differences in terms of a significant number of new hardware, the installation of a new complex software, or the emergence of new requirements that are not implemented as part of the implementation of previous decisions, the creation of such a hardware and software package will be a project. Those. The project executor always is in the beginning of its work in a state when the goals are determined, the solutions are undefined, the successful solution of the task is to. I explain why I stopped in detail, it would seem that the terminological question.

The fact is that there are 2 approaches to the performance of work and their assessment. This is a developer approach and the customer's approach.

The developer is trying when the task is to be implemented:

1. Try to apply the decision previously implemented by the previously implemented;
2. In case of the impossibility of this, it is trying to apply the decision-tested by other firms (most often a solution recommended by the manufacturer of equipment or software);
3. Try to reduce the requirements of the customer and, if possible, to reduce them to the same model solutions;
4. In case of failure of the previous paragraph, the developer is trying to increase the time of work or make softer requirements for acceptance of its work;
5. Try at the reception stage to concentrate on strengths completed project and hide your mistakes and not refinement;
6. Try to quickly pass the project and start a new one, or, as a last resort, to secure outsourcing.

The customer's approach is primarily characterized by:

1. An attempt to get as much from the developer and for smaller money;
2. Attempts in the process of developing a project to change or clarify the items of the initial TK;
3. During the reception, try to get as much documentation as possible, and find developer errors;
4. To try at the expense of the Customer not only to correct the errors identified during acceptance process, but also make the next changes to the project.

Therefore, the use of implementation, instead of developing a significantly less chance of the project's success - is always preferably preferably for the Contractor. The above option is of course the most relevant, if the development of the project leads a third-party organization. In fact, when ordering a truly complex project (and the construction of a data center is to such projects and belongs to a third-party company, the participation of customer specialists is absolutely necessary at the initial stages of the project. Indeed, no one knows the requirements for the created data center, as the customer's specialists. Of course the customer, at a minimum should be able to control the execution of the project, it is more accurate to have information about the timing of each stage, the course of its execution, as well as not just participate in the project acceptance, but also to participate in writing testing programs. Only in this case there is a fairly accurate wording of those. tasks, operational solution arising issues, a comprehensive verification of the result obtained.

There are two options for the execution of the project to build a data center. The first implies the implementation of the project on its own, and the second imposes these duties on a third-party artist. In pure form such schemes are rare. Almost always building such systems joint work of the artist (or several performers) and the customer. But everything rests on the question of who will lead the project. It would seem that to whom, as not to the Contractor to give such rights, but ... participation in writing the TK at the same time the customer (since he knows all the requirements for his data center) so the performer (because if not to attract the performer, then the customer may well write Such a TK, which no one can implement at all) allows you to work out in the process of discussion a fairly accurate idea of \u200b\u200bthe system that will be created, and softwarethat should be applied. Those. Specialists involved in writing TK are at the time of its expiration the most competent In terms of specific requirements for the project executed for a specific customer. Immediately answer possible questions about joint writing TK. Customer in the development of large projects may alone write only preliminary TK, suitable as possible only for the competition When searching for artist. And jointly written TK with controversial issues with the customer with controversial issues will serve as a major document in the receiving the CDA, since the "Program and Test Methodology" will be written on the basis of TK.

Therefore, one of the main errors of the customer is the elimination of specialists of the participating TK and episodic participation in the sketch and work project of only narrow specialists in solving private issues. Specialists involved in the work on the implementation of large projects should the customer be in the department of comprehensive work. And it is they who must attract if necessary for all specialists in certain areas. In this case, the Specialists of the Complex Department will be aware of all the "thin" project sites and the project itself will receive great chances of successful completion. Also, the specialists of the integrated department should participate in the acceptance of the customer's work, because Constantly following the work of the work, they will be aware of all his problems.

Remark on the work of the work appropriate.

It is incorrect to think that the download of the integrated department will be limited only to participation in large projects, which the customer is usually not very much. Large projects exist in themselves. Usually each project requires its expansion, docking with various subsystems, making changes to the newly emerging tasks. It is in solving these issues and use the complexes-complexes. The previous one concerned not only large projects, because it is necessary to understand that only the introduction of individual products not affecting a large number of customer employees, it is possible to implement failing a comprehensive department.

If we turn to the experience of implementing large projects, we note that large organizations (for example, banks), or those whose specialization is associated with IT, they themselves manage projects for the creation of their data center.

Summing up the stages of the rationale and the compilation of TK

From the above, we can conclude:

1. Speaking about creating a data center, you must first arrange the priority of the requirements that it will have to satisfy.
2. After placement of priorities, you must take the basis of one of the standards whose requirements you will follow. (I would advise using TIA-942.But it is impossible to forget that he does not consider the issues of operation.)
3. All retreats from the standard for the better or worse must be reasonable.
4. To compile the TK, it is necessary to use its comprehensive work department (or create it), because You need people personally interested in the successful implementation of the project and which will oversee all work with the performer.

If you notice that in this part I considered questions before the start of writing TK, stressed that it is necessary to write a TK with the Contractor, and nothing wrote anything about the choice of the Contractor. The fact is that the choice of the performer is a separate and responsible task. And if it is very briefly mentioned, then usually the choice is divided into 2 stages:

1. Definition of the Circle Circle to solve the problem of building your specific data center.
2. Analysis of the material represented by firms and clarifying issues at personal meetings.

It is usually easier by choosing several firms implementing successful projects in this area to provide them with preliminary TK (such a TK is possible to draw up by the artist's specialists). Then the candidates for building a data center are asked to make a small document, in briefly describing all the subsystems of the data center and the process of its operation. Usually, by the completeness of the issues under consideration, the validity of solutions and the results of personal communication, the choice of the Contractor becomes obvious. And I will add from myself: if you are promised at a personal meeting and for cheap (in any case, it is significantly cheaper than others) this is a reason not to believe and check the reality and the quality of the projects performed by the company. In addition, often in truly complex projects of building a data center, the execution of some kind of subsystems requires the involvement of other firms. In this case, immediately need to negotiate that one of the firms is for this project. system integrator And all technical and other questions you will deal with it. There is nothing worse than the "piece" project implementation. And then, with any trouble, everything will be as an immortal monologue of Rykin "to the buttons there are a claim?".

The era of computers has been more than 50 years old, respectively, the infrastructure of their livelihoods is as much. The first computer systems were very complex in use and maintenance, they required for the functioning of a special integrated infrastructure.

The incredible set of cables combined the various subsystems, and many technical solutions were developed for their organization applied until now: racks for equipment, false floors, cable trays, etc. In addition, cooling systems were required to prevent overheating. And since the first computers were military, security issues and limitations were in the first places. Further, computers have become less, cheaper, unpretentious and penetrated into a variety of industries. At the same time, the need for infrastructure disappeared, and computers began to place anywhere.

The revolution occurred in the 90s, after the client-server model distributed. Those computers that began to count servers began to be placed in separate rooms with the prepared infrastructure. The names of these premises in English sounded like Computer Room, Server Room, Data Center, while in the Soviet Union we called them "machine rooms" or "computing centers". After the collapse of the USSR and the promotion of English terminology, our MC turned into "server" and "data centers" (data center). Are there any fundamental differences between these concepts, or is it just a matter of terminology?

The first thing that comes to mind is the scale: if small, then the server, and if large is the data center. Or: if inside only your servers, then this is a server; And if server placement services are provided to third-party companies - then the data center. Is it so? Over the answer we turn to standards.

Standards and criteria

The most common standard, describing the device of data centers, is American TIA 942. The Russian counterpart, unfortunately, no, Soviet CH 512-78 has long been and hopelessly outdated (albeit from 2000), it can only be considered From the point of view of common approaches.

In the Standard TIA 942 itself, it is written that the purpose of its creation is to formulate the requirements and guidelines for the design and installation of the data center or the machine room. We assume that the Code is what matches the requirements of TIA 942, and the server is just a certain room with servers.

So, the standard TIA 942 classifies 4 levels (Tiers) of the data center and calls a number of parameters for which this classification can be carried out. For example, I decided to check whether my server, built together with the factory three years ago, a real data center.

As a small digression, I will indicate that the plant is engaged in the manufacture of stamped parts for the automotive industry. We produce body details for companies such as Ford and GM. The company itself is small (general staff of about 150 people), but with a very high level of automation: the number of robots is comparable to the number of workers in the workshop. The main difference of our production can be called the rhythm of the work of Just-in-Time, that is, we cannot afford the delay, including the fault of IT systems. IT are critical for business.

The server was designed to ensure the needs of the plant, to provide services to third-party companies were not supposed, respectively, and certification for compliance with any standards was not required. However, since our plant is a member of a major international holding, design and construction went taking into account the internal corporate standards. And these standards are at least partially built on the basis of international.

The standard TIA 942 is very extensive and describes in detail the approaches to the design and construction of the data center. In addition, in the annex to it there is a large table with more than two parameters for compliance with the four levels of the data center. Naturally, they are inappropriate to consider them all in the context of this topic, and some of them, for example, such as "separate parking for visitors and employees", "The thickness of the concrete slab at the ground level" and "proximity to airports" are not very direct attitude to the classification Cododas and especially difference from the server. Therefore, we will only consider the most important, in my opinion, parameters.

Main parameters of the classification of the data center

The standard establishes the criteria of two categories - mandatory and recommended. Mandatory designated in the word "must" (SHALL), recommended - words "should", "can", "preferably" (Should, May, DesiRable).
The first and most important criterion is the level of operational readiness. According to TIA 942, the highest-fourth data center - a level should have 99.995% of readiness (ie, no more than 15 minutes of idle per year). Further, on the downward, 99.982%, 99.749% and 99.671% for the first level, which corresponds already to 28 hours of idle per year. The criteria are rather tough, however, what is the operational readiness of the Data Code? It is considered only the simple entire data center in the fault of one of the livelihood systems, and simple separate servers for the operational readiness of the CDA does not affect. And if so, then the most likely cause of refusal to fairly consider interruptions in the energy supply system.

In our server installed a powerful APC UPS with reservation N + 1 and an additional battery cabinet that is capable of supporting the performance of not only servers, but also all computers in the enterprise up to 7 hours (and why we need working servers, if you have no one to connect to them). For three years of operation, no failures were never, so on this parameter we can claim the highest Tier 4.

By the word about the power supply, the third and fourth CDA classes require second energy. We do not have it, so the maximum is the second class. Still standard classifies power consumption per square meter area. The strange parameter, never thought about it. I stumbled: I have 6 kW per 20 sq. M., that is, 300 W per square meter (only the first level). Although it is possible that I misunderstand: The standard states that a good Code must have free areas for scaling. That is, it turns out, the more "scaling supply", the lower the level of the central level, and should be the opposite. Here we have the lowest estimate, but still the standard is responsible.

For me, an important parameter is an assembly assembly of external telecommunication systems. We carry out online interaction with clients for receiving orders and shipment of components, respectively, the lack of communication can lead to the conveyor of our customers. And this not only negatively affects our reputation, but will lead to serious fines. Interestingly, in the standard it is stated on the duplication of the input points of communication, and there is nothing in the annex (although it is indicated that in the levels above the first all subsystems should be redundant). We use two channels for connecting automatic routing in case of refusal in one of them, plus a backup GPRS router with manual connections. Here we again meet the highest requirements.

A significant part of the standard is dedicated to cable networks and systems. These are the distribution points of the main and vertical subsystems of the general cable data center and the cable layout infrastructure. After reading several parts of this section, I realized that it should be either learned by heart, or to compete and concentrate on more important things. Although on a superficial view (category 6 twisted pair, separation of active equipment from passive) we still comply with the standard. Although I'm not sure in such parameters as the distance between the cabinets, the bending angles of trays and the correctness of the separation of the tracks for low-current cables, optics and power. We assume that here we are responsible for the requirements.

Air conditioning systems: There are air conditioners, there is a reservation, it can be said that there is even a cold and hot corridor (although one, due to the size of the room.) But cooling is not distributed under false floors, as recommended, but directly in the working area. Well, we do not control the humidity, and according to the standard it is omission. We put partial compliance.

Separate part is devoted to false floors. The standard regulates both the height and the load on them. Moreover, the higher the class center, the higher and more powerful should be false floors. We have them, and in height and loads correspond to the second class of the data center. But my opinion is that the presence of false floors should not be a criterion, and even more than the characteristic of the central system. I was in the data center of the company "Westcall", where they initially refused the false floors, placing all trays under the ceiling. Air conditioning is made with cold and hot corridors. The building is detached, the room is large, provide specific services. That is, a good, "real" data center, and it turns out that without false floors, it does not correspond to the standard.
Following important moment - safety system. Great data centers are guarded by almost like safe cells in a bank, and to get there - a whole procedure, ranging from coordination at different levels and ending with dressing up and bugs. We have easier with this, but everything is present: physical security provides chop, which protects the factory itself, and the access control system ensures that only authorized employees fall into the premises. We put a plus sign.

And finally, the gas fire extinguishing system. The main and backup cylinders, sensors in the indoor indoor, under the floor and above the ceiling and control system - everything is there. By the way, an interesting moment. When companies want to boast their data center, first of all show a fire extinguishing system. Probably because this is the most unusual element of the data center, not occurring practically anywhere, except for the dwarves, and the rest of the equipment simply looks like closets of different color and size.

The main thing, in my opinion, the difference between two upper levels Code from lower - what they should be located in a separate building. It would seem that there is a sacral meaning of the difference between the server from the data center: if it is highlighted in a separate building, then this is a data center. But no, the standard says that the first two levels are also the centries.

I found the parameter for which my server on the data center does not pull: the size of the front door. According to the standard there must be a minimum of 1.0 × 2.13 m, and better 1.2 × 2.13 m. And we have an ordinary door: 0.9 × 2.0 m. It is minus, but consider the criterion for the differences in the server from the server The size of the front door is not serious.

Almost real data center!

So, what did we do? The small server at the factory corresponds to almost all the requirements of the standard for the organization of the data center, even if with small reservations. The only serious inconsistency is the size of the entrance door. The absence of a separate building under the server does not leave no chance of higher space. It means that the assumption that the data center is necessarily large, and the server, on the contrary, is always small, incorrectly. As well as the second assumption that the Code serves many customer companies. It follows that the server is only synonymous with the data center.

The concept of the data center appeared when they began to sell hosting services, rental racks and server placement. At that time, the concept of the server was devalued by a negligence attitude to the infrastructure due to the unpretentiousness of the PC and the low price of downtime. And in order to show that the provider is all built for comfortable and trouble-free work, and they are able to guarantee the quality of the service, the concept of the data center introduced, and then the standards of their construction. Given the trends in centralization, globalization and virtualization, I think that the concept of the server will soon disappear or turn into the designation of the telecommunications node.

I suppose, about the same it is calculated by our president with the law on the police. The concept of "militia" was devalued, and to create new rules for them is too late. Whether it will be possible to build literate standards for a new structure - let's see in the near future.