DATA CENTERS EVOLUTION & ARCHITECTURES

WHAT ARE DATA CENTERS?

Data centers first emerged in the early 1940s when computer hardware was heavy and too complex to operate and maintain. Early computer systems required many large components that operators had to connect with many cables. They also consumed a large amount of power and required cooling to prevent overheating. To manage these computers, called mainframes, companies typically placed all the hardware in a single room, called a data center. Every company invested in and maintained its own data center facility.

HOW HAVE DATA CENTERS EVOLVED?

Data centers have evolved significantly in recent years as enterprise IT needs continue to move toward on-demand services. There is an expression these days: the modern data center is where your workloads are.

To support this level of application elasticity and mobility, enterprises are transforming their data centers with a modern architecture. A modern data center relies on virtualization, cloud, and software-defined networking to deliver application workloads everywhere; this includes physical data centers and both multicloud and hybrid environments.

A modern infrastructure allows your organization to extend into cloud services. This evolution enables flexible scaling for network, storage, and compute demand surges. Modern data centers are very different than they were just a short time ago. Infrastructure has shifted from traditional on-premises physical servers to virtual networks that support applications and workloads across pools of physical infrastructure and into a multicloud environment. Over time, innovations in hardware technology reduced the size and power requirements of computers. However, at the same time, IT systems became more complex, such as in the following ways:

• The amount of data generated and stored by companies increased exponentially.
• Virtualization technology separated software from the underlying hardware.
• Innovations in networking made it possible to run applications on remote hardware.

Modern data center design evolved to better manage IT complexity. Companies used data centers to store physical infrastructure in a central location that they could access from anywhere. With the emergence of cloud computing, third-party companies manage and maintain data centers and offer infrastructure as a service to other organizations.

SO, WHAT IS WRAPPED WITHIN THESE DATA CENTERS?

Most enterprise data center infrastructure falls into three broad categories:

• Compute
• Storage
• Network

Also, data center equipment includes support infrastructure like power systems, which help the main equipment function effectively.

COMPUTING INFRASTRUCTURE

Computing resources include several types of servers with varying internal memory, processing power, and other specifications as below:

• Rack Servers: Rack servers have a flat, rectangular design, and you can stack them in racks or shelves in a server cabinet. The cabinet has special features like mesh doors, sliding shelves, and space for other data center resources like cables and fans.
• Blade Servers: A blade server is a modular device, and you can stack multiple servers in a smaller area. The server itself is physically thin and typically only has memory, CPUs, integrated network controllers, and some built-in storage drives. You can slide multiple servers into a storage unit called a chassis. The chassis facilitates any additional components that the servers inside it require. Blade servers take up less space than rack servers and offer higher processing speed, minimal wiring, and lower power consumption.

STORAGE INFRASTRUCTURE

The following are two types of data center storage systems:

• Block Storage Devices: Block storage devices like hard drives and solid-state drives store data in blocks and provide many terabytes of data capacity. Storage Area Networks (SANs) are storage units that contain several internal drives and act as large block storage systems.
• File Storage Devices: File storage devices, like Network-Attached Storage (NAS), can store a large volume of files. You can use them to create image and video archives.

NETWORK INFRASTRUCTURE

A large number of networking devices, such as cables, switches, routers, and firewalls, connect other data center components to each other and to end-user locations. They provide flawless data movement and connectivity across the system.

SUPPORT INFRASTRUCTURE

Data centers also contain these components:

• Power Subsystems
• Uninterruptible Power Supplies (UPS)
• Backup Generators
• Ventilation and Cooling Equipment
• Fire Suppression Systems
• Building Security Systems

These data center components support the main equipment so that you can use the data center facilities without interruption.

WHAT ARE THE TYPES OF DATA CENTERS?

As of today, we do have different types of data centers and service models available. Their classification depends on whether they are owned by one or many organizations, how they fit (if they fit) into the topology of other data centers, what technologies they use for computing and storage, and even their energy efficiency.

There are four types of data centers:

1. ENTERPRISE DATA CENTERS : These are built, owned, and operated by companies and are optimized for their end users. Most often they are housed on the corporate campus.
2. MANAGED SERVICES DATA CENTERS : These data centers are managed by a third party (or a managed services provider) on behalf of a company. The company leases the equipment and infrastructure instead of buying it.
3. COLOCATION DATA CENTERS : In colocation ("colo") data centers, a company rents space within a data center owned by others and located off company premises. The colocation data center hosts the infrastructure: building, cooling, bandwidth, security, etc., while the company provides and manages the components, including servers, storage, and firewalls.
4. CLOUD DATA CENTERS : In this off-premises form of data center, data and applications are hosted by a cloud services provider such as Amazon Web Services (AWS), Microsoft (Azure), or IBM Cloud or other public cloud provider.

WHAT ARE THE STANDARDS IN DATA CENTER DESIGN?

As data centers increased in size and complexity and began to store sensitive and critical information, governments and other organizations imposed regulations on them. The Telecommunications Industry Association (TIA) established four levels or standards that cover all aspects of data center design, including:

• Architecture and Topology
• Environmental Design
• Power and Cooling Systems and Distribution
• Cabling Systems, Pathways, and Redundancy
• Safety and Physical Security

Similarly, the Uptime Institute established four tiers to compare site performance objectively and align infrastructure investments to business goals. We list the four data center tiers below.

1. TIER I data center is the basic capacity level to support IT systems for an office setting and beyond. Some of the requirements for a Tier I facility include:

• Uninterruptible Power Supply (UPS) for power outages and spikes
• A physical area for IT systems
• Dedicated cooling equipment that runs 24/7
• A backup power generator

Tier I protects against service disruptions from human error but not against unexpected failure or outage. You can also expect an annual downtime of 29 hours in Tier I data centers.

2. TIER II facilities provide additional cooling components for better maintenance and safety against disruptions. For example, these data centers must have the following:

• Engine generators
• Chillers
• Cooling units
• Pumps

Although you can remove components from Tier II data centers without shutting them down, unexpected failures can affect the system. You can expect an annual downtime of 22 hours from a Tier II data center.

3. TIER III data centers provide greater data redundancy, and you can maintain or replace equipment without system shutdown. They also implement redundancy on support systems like power and cooling units to guarantee only 1.6 hours of annual downtime.

4. TIER IV data centers contain several physically isolated systems to avoid disruption from both planned and unplanned events. They are completely fault-tolerant with fully redundant systems and can guarantee a downtime of only 26 minutes each year.

CLOSING NOTES

There has been a wave of change in the way data centers are being designed today, and this approach is majorly driven by the trends of cloud computing and advanced storage mechanisms.

As we see today, the on-premise data centers are being moved to hybrid cloud models which enables the enterprises to not only keep their security & compliance requirements intact but also allows them to bring down the IT infrastructure & operations costs.

We also have a heap of GenAI tools being developed by major IT creators which would further change the data center ecosystem in the near future. As GenAI creates a massive shift in the global data center landscape, the critical trends are emerging.

Over the past year, the global data center industry has witnessed the start of a truly transformative era accelerated by the hyper ascent of Generative Artificial Intelligence (GenAI). It is becoming increasingly critical to prepare existing data center resources for this GenAI tidal wave to ensure a seamless integration that aligns with organizational goals. However, before deploying GenAI, IT leaders must address foundational tasks such as identifying a suitable model, gathering relevant data, and training the model for inference. However, organizations in highly regulated industries like finance and healthcare must consider the infrastructure implications of deploying GenAI. Integrating GenAI into data centers is not just a technological upgrade; it is a strategic business decision. In 2024, data center managers must navigate this new environment by intelligently modernizing IT infrastructure without compromising security and controlling costs. By doing so, organizations can effectively meet the rising demands of GenAI while leveraging its potential for driving business growth and efficiency.

Let's continue to adapt, innovate, and collaborate, ensuring that our data centers remain at the forefront of driving business excellence in the digital age.