The core of a reliable information technology system is the server room, designed and implemented in a professional manner. In addition to the architectural structure designed for this special purpose, the operations of a number of weak and heavy current systems should be ensured and harmonized in an extremely reliable manner.

These subsystems are:

• redundant and precision air conditioning
• redundant uninterrupted power supply, diesel power generator
• fire alarm equipment and fire extinguishers
• access control system
• rack control system
• planning and implementing internal data network, external optical and copper cable connections
• power surge and EMC protection

The information available to us today and the reliable operation of IT applications critical to business success. At the same time, not everyone is aware that for the servers, storing precious information such operating conditions should be ensured which are totally different from those of the PCs of normal users. The requirements for appropriate construction of the equipment rooms are specified in standard TIA-942. The objective of the standard is to provide assistance and specify certain requirements for planning an up-to-date datacenter (equipment room).

This standard classifies the equipment rooms into 4 different categories based on their applicability and security. The classification of some components of the equipment room (walls, air conditioning, UPS, telecommunication distributors, etc.) may be different, but in this case the datacenter is to be classified into the category corresponding to the “weakest link”.

Tier 1 Datacenter - "basic"

The first class meets the basic requirements and has no redundancy at all. The built-in devices are functioning separately. Any breakdown may lead to an immediate collapse of the equipment room.

Tier 2 Datacenter - "redundant components"

The equipment rooms having redundant technological elements (air conditioning, UPS, etc.) belong to this class, but they are operated by a single feeding circuit or, for air conditioning, by a single outdoor piping. Disasters deriving from possible breakdowns may be ruled out more efficiently in this system. To sustain performance some elements require regular restarts.

Tier 3 Datacenter - "simultaneously maintainable"

The Tier 3 datacenter has several feeding and cooling circuits, but only one is active at a time. Since redundant equipment is not integrated with theactive circuits, the operation of the room can also be maintained in case of system failures. The capability of a Tier 3 level allows accomplishing each planned activity (test, maintenance) without disturbing the operation of the computers. On the other hand, non-planned events and operational failures may lead to the collapse of the equipment room. Human assistance is needed to change circuits therefore the room requires 24-hour surveillance.

Tier 4 Datacenter - "error tolerant"

The difference between rooms classified into this category and the previous class is that here 2 circuits are active at all times. The integrated IT devices are designed for a very high availability, "CPU clustering" and "Redundant Array of Independent Disk/Direct Access Storage Device (RAID/DASD)" technologies are typical designs. In the equipment room, all planned activities can be accomplished and, owing to several circuits operated simultaneously, the system will work even in case of disturbance deriving from some failure.

Parameters of different equipment rooms are shown below:

 

This classification is only one of the requirements specified by the standard TIA-942 to be considered for planning an equipment room. For establishing a datacenter with a high reliability, we may only achieve our objective, i.e. establishing physical security for our data, by selecting a company having the appropriate skills, professional competence and being familiar with all the details of the applicable standards.

 

WHY SYNERGON?

An equipment room planned appropriately and built to achieve the highest quality provides a good rate of return-on-investment. The return on investment is closely related to the data security provided by it.

For the servers of the applications critical to the operation of the company, operating conditions providing the highest possible operational reliability should be ensured. Operating troubles may be caused by elevated temperatures, over-voltage impulses or power supply outages.

The repair fees of the servers and auxiliary systems which failed due to the aforementioned problems usually amount to minor costs only, though these devices are very expensive. Lost data restoration is much more expensive, if it is possible at all.
The costs of the lost operating time may also be of greater magnitude when a large number of employees are hindered from performing their daily tasks, and direct business actions may come to a stand still as a result of such breakdowns.

 

OUR MAIN REFERENCES

• EDS Hungary Ltd.
• EGIS Pharmaceuticals Plc.
• Employment Office
• Raiffeisen Leasing
• Tata Consultancy Services Limited – Hungarian Branch
• Invitel DataCenter