Data Center Refers To A Physical Place Organizations Use To Host Applications And Important Data. If They Are Engaged In Providing Cloud Services, They Provide Services To Their Customers. 

The design of a data center, which consists of computing resources and storage, allows organizations to provide applications and data in a shared form and through access rights to customers or employees.

Typically, a data center includes routers, switches, firewalls, storage, servers, Ethernet or fiber optic cabling, applications, and tools needed to access network resources and networked applications.

What are the characteristics of a modern data center?

Modern data centers have had significant differences compared to the past, as the infrastructure has moved from traditional on-premise servers to virtual networks. In the new architecture, applications and data traffic are deployed in physical infrastructures based on multi-cloud services. An important principle that should be paid attention to in this context is establishing a balance between private and public cloud services because data centers are used to communicate between different sites. These sites may be located near the company or far away. It should be explained that public cloud infrastructure is also a collection of several connected data centers.

Why are data centers important for organizations and businesses?

Information technology data centers are designed to run business applications and simplify access to computing and storage resources. Among the essential programs in this field, the following should be mentioned:

•  Implementation of email servers and file sharing
•  Hosting large applications such as databases
•  Deployment of customer relationship management software (CRM)
•  Establishing programs related to enterprise resource planning (ERP) and databases
•  Big data processing and analysis to use in machine learning and business intelligence
•  Implementation of virtual machines, communication, and collaboration services

Components of data centers

As mentioned, the data center design is based on deploying routers, switches, firewalls, storage, servers, and application access tools. While hardware components play a significant role in implementing a data center, the topologies, and schemas allow these hardware components to create a unified platform called a data center. 

These components are as follows:

• Network infrastructure: physical and virtual servers, data center services, storage space, and external connections that enable users to connect with the data center.
• Storage infrastructure: Data centers are implemented to store data; for this reason, they consist of storage and storage area networks (SAN) that enable the process of hosting and quick access to information. In small data centers, storage is connected or servers in the form of arrays or clusters, while on a large scale, a separate network from the local network is created for them to communicate between clients and the web through mechanisms such as Fiber Channels. to establish
• Computing resources: servers responsible for receiving, processing, and responding to client requests. Typically, network administrators use rackmount servers when implementing data centers. Still, if the field of activity is a large enterprise, blade servers are used so that each blade is responsible for performing a specific task.
• Applications: They are in charge of accessing computing and storage resources. As mentioned, servers are responsible for processing data, while applications provide access to data hosted on the server.
• Redundancy: Typically, network administrators design data centers based on the principle of redundancy of hardware and communication links to protect data center services and continue business activities.
• Security equipment: Network security tools include firewalls, intrusion detection, and prevention tools to ensure the data center’s security.
• Ensuring the delivery and performance of programs: includes mechanisms to maintain the performance of programs, the flexibility and availability of programs even in the event of failure, and increased traffic.

What do data center planning and design mean?

Data center design designs and models data center system architecture, information technology resources, and infrastructure. This planning refers to the logic of the data center, defined before development or deployment in the organization’s IT environment.

Data centers with a structured and integrated design allow business operations to run smoothly and simplify the data management process. Data center design can be done using the organization’s IT staff or by hiring a service provider. Typically, data centers require reliable power sources, regular data backups, reliable and stable communication paths for data access, environmental control, and security measures. For this reason, it is essential to have a data center planning process that provides experts with a clear vision of what should be done. In the planning process, network experts deal with the following topics:

• They choose the type of modular data center construction process.
•  They provide a list of required application software and specify the type of server operating system to be installed.
•  They find ways to integrate operations.
•  If possible, they share resources.
•  They consider the deployment of cloud computing.
•  A business continuity plan considers data and business operations.

You can see the design pattern of a typical data center in Figure 1.

figure 1

Data center design standards and guidelines

Data center design standards define minimum requirements and best practices when designing, developing, and maintaining data center systems. Professional organizations such as the International Organization for Standardization (ISO) or the Telecommunications Industry Association have provided guidelines for effectively designing and managing data centers. In general, the design standards that should be considered  when building a data center are as follows:

•  Datacenter telecommunications infrastructure standard (TIA-942)
•  ISO / IEC 27001:2005 and 27001:2013 information security management system standard

These standards specify the requirements for the successful design and deployment of a facility and provide guidelines for the following concepts:

•  Physical regulations for entering and exiting the data center
•  Information security
•  Design and flexibility
•  Providing physical and virtual infrastructure security
•  Optimal use of energy
•  The need to pay attention to backup plans for business continuity
• Compliance with data center  design best practices
•  The need to pay attention to what the data center is supposed to do; To achieve the best performance
• Attention to data center  energy supply and emergency power

A big problem when designing a data center is the financial costs of implementing such a design.

More specifically, regarding data center design, servers, cooling systems, emergency power, and the physical space where the equipment will be located increase costs significantly.

For example, experts in the physical space department should pay special attention to the floor space, cabinets and shelves, location, and how to install the cooling systems.

For this reason, to reduce costs, some organizations try to implement data centers only to respond to current needs. Since data centers host expensive equipment that works almost 24 hours daily, they need a proper cooling system.

This cooling system must be able to circulate air to remove heat from the equipment and keep the equipment as cool as possible. Typically, network experts use hot and cold corridors for this purpose (Figure 2).

figure 2

Security

Security is the most crucial element in the design of a data center that should not be ignored. Network security, data encryption, and data backup are essential issues when building a data center.

Mistakes you should not make in data center design

Setting up a data center means using the latest technologies so that the costs do not increase too much. Unfortunately, today some companies seek to establish data centers that meet short-term needs so that the company faces various bottlenecks in the long term. The critical thing to remember is that data is constantly growing, so building a future-proof data center system that can upgrade and has room to deploy new equipment is essential.

Statistics show that network experts make the following nine mistakes when designing data centers:

•  Failure to conduct a full infrastructure assessment
•  Miscalculating Total Cost of Ownership (TCO)
•  Having a project manager who is unable to see the details well.
•  Start deployment before final planning is complete.
• Wrong choice of data center  power supply equipment
•  Failure to identify dependencies
• Wrong formulation of data center  design timeline
•  Misalignment of the data center with business objectives

What services do data centers offer?

Typically, data centers provide businesses with the following features:

•  Providing web hosting services such as web service, database, domain name system, and file sharing through FTP server
•  Providing services for email services
•  Providing services for hosting dedicated and shared servers (Colocation and Dedicated Servers)
•  Providing services for applications (CRM, ERP, and similar examples)
•  Providing services for online network games

Data center infrastructure standards

The most common standard used in the design of data centers and their infrastructure is the ANSI/TIA-942 standard. This standard includes issuing ANSI/TIA-942 compliant certifications that ensure data center compliance and compatibility with one of the four main groups of data centers. It ensures that the data center has acceptable stability and fault tolerance.

Typically, data centers, technically and internationally, belong to one of the following four main groups:

• Basic site infrastructure Tier 1: A data center with the standard level of the first layer (Tier 1) includes limited protection against physical accidents. This level has single-capacity components and a non-redundant distribution path.
• Redundant-capacity component site infrastructure Tier 2: The data center with the standard tier two level (Tier 2) provides more protection against physical incidents. This standard includes Redundant-Capacity components and a non-redundant distributed path.
• Concurrently maintainable site infrastructure Tier 3: This data center is based on applying the best policies to deal with physical and virtual incidents, includes equipment redundancy, and has several independent distribution paths. The advantage of this data center is that if any component fails, it can be replaced without disrupting the service. For example, if the power supply of a switch fails, a secondary power supply is introduced into the circuit. In this case, the network experts have enough time to replace the damaged power supply.
• Fault-tolerant site infrastructure Tier 4: The highest standard in the field of data center design, which provides the highest level of fault tolerance and redundancy. This standard emphasizes equipment redundancy, multiple independent distribution paths, concurrent support, troubleshooting, and providing the highest level of physical and virtual security. Financial institutions, military and government centers typically use these data centers due to their high cost.

How many groups are data centers divided into?

In addition to the four main groups we mentioned, data centers are divided into different types, each providing specific services. In the discussion of data center classification, there are various questions. For example, does the data center belong to one organization, or do several organizations use it? What is the layout and topology of the data center design, what technologies are used in computing and storage, and what arrangements are made for proper energy efficiency? It has been used. In terms of functionality and design, data centers belong to one of the following four groups:

Enterprise data centers

These data centers are set up, managed, and controlled by the companies themselves and are designed to meet the needs of the employees of an organization. In this case, a team of IT experts is responsible for maintaining and optimizing the performance of the data center.

Managed data centers

Third-party companies or data center service providers manage these data centers. In fact, instead of buying equipment and infrastructure, customers rent these data centers from service providers.

Colocation data centers

In colocation data centers, the customer rents space from a third-party company that owns its data centers. Colocation data centers are responsible for hosting and maintaining infrastructure such as buildings, cooling systems, bandwidth, security, and power supply.

In this case, the customer uses the services the third party provides but is responsible for maintaining the servers, storage space, and firewalls.

Cloud data centers

In this data center model, data and applications are hosted by a cloud services provider such as Amazon Web Services (AWS), Microsoft (Azure), IBM Cloud, or other public cloud service providers. There are similar data centers in Iran.

Data center design

As you have seen, today, there are different types of data centers, each of which consists of other equipment designed for specific applications. For this reason, the design of a data center must be done according to standards.

When you plan to design a data center, it is better to pay attention to the following points to have a structured and accurate design:

•  Preparing a preliminary design and preparing technical drawings
•  Obtaining the necessary permits
•  Finalizing the initial plan, which includes the selection of equipment (servers, switches, switches, etc.), the type of cabling, and emergency power supply equipment (e.g., generators, UPS, etc.).
•  Preparation of documents
•  Design cost estimation
•  Determine the final budget
•  Evaluation of the organization and the physical location where the equipment is to be installed.
•  Ordering, buying, and transporting equipment to the place
•  Provision of communication links and telephone lines
•  Preparation of design plans (plans related to cabling inside the racks, type of architecture and topology used in the design, cabling operations in the building, control and access, fire extinguishing, emergency power supply plan, generator, power, and data network installation)
•  Preparing the implementation and support team
•  Installation and commissioning of equipment
•  Initial tests
•  The final setup of the data center

last word

The future of data centers is bright because various companies provide data center services to businesses and industries in this field. Data centers offer the converged infrastructure that appears to be moving toward greater convergence with the cloud. An essential point in this field is security.

Many companies believe that based on a detailed design, most risks can be identified and eliminated before they occur. So that business activities continue and income levels increase.

Based on the converged data center architecture, a single interface can be used to manage the entire business infrastructure and simplify each process, which increases data security and better availability.

Also, the performance of virtual solutions and the convergence of all operational layers, such as computing, network, and storage, significantly improve the data center.