Data Center Infrastructure Resilience: The Four-Tier Classification System

Data centers are not all created equal. In fact, there are several different types of data centers, each one offering specific features and capabilities. To be able to compare different data centers, the industry employs a four-tier classification system created by the Uptime Institute. This system assesses the level of redundancy and potential downtime in a year. Each data center is assigned a tier certification based on criteria such as power, cooling, maintenance protocols, redundancy and fault tolerance.

Tier 1 Data Center – Basic Capacity:

Tier 1 data centers provide basic infrastructure capacity with no redundancy. While they provide essential services, the data center may require maintenance or equipment upgrades on occasion. These data centers are appropriate for small businesses or non-critical applications that can withstand brief outages.

Tier 2 Data Center – Redundant Capacity:

Redundant components are used in Tier 2 data centers to improve reliability. They may, however, experience planned downtime for maintenance purposes. These Tier 2 data centers are a step up from Tier 1 data centers and are appropriate for businesses that require improved availability and reliability without the need for continuous uptime.

Tier 3 Data Center – Concurrently Maintainable:

A Tier 3 data center is designed with concurrent maintainability in mind. It distinguishes itself by having redundant components and distribution paths to serve the critical environment. This means that maintenance or replacement of equipment can be performed without any need for shutdowns. In Tier 3, the additional components are integrated with the Tier 2 setup, ensuring that any part can be shut down without impacting the IT operations.

Tier 4 Data Center – Fault Tolerant:

Tier 4 data centers take redundancy to the next level. It consists of multiple independent systems that are physically isolated from each other. These systems act as redundant capacity components and distribution paths. The purpose of this separation is to prevent a single event from compromising both systems. In a Tier 4 environment, disruptions from planned or unplanned events will not affect the overall operation. However, if the redundant components or distribution paths require maintenance and are shut down, there is a slightly higher risk of disruption if a failure were to occur.

In a Tier 4 facility, fault tolerance is a key feature added to the Tier 3 topology. This means that even if a piece of equipment fails or there is an interruption in the distribution path, critical IT operations will continue to function without any impact. To achieve this level of fault tolerance, all IT equipment within the data center must have a fault-tolerant power design in place. Additionally, Tier 4 data centers place a high emphasis on continuous cooling to maintain a stable and optimal environment for the equipment housed within.

When evaluating data center options, businesses must carefully consider their specific needs, budget, and goals. Choosing the appropriate data center tier ensures that their critical applications and services are resilient and available.

Choosing the Right Data Center Strategy: Comparing Enterprise and Colocation Models

When choosing a data center strategy, it is crucial to compare enterprise data centers and colocation data centers to determine the best fit for your organization’s needs.

Enterprise Data Centers:

Enterprise data centers are privately owned and operated facilities dedicated to meeting the unique IT infrastructure needs of an organization. These data centers are an ideal choice for organizations seeking customized networks or benefiting from economies of scale due to the substantial amount of traffic or data being processed and managed within their infrastructure.

There are various reasons why organizations opt to maintain enterprise data centers. These include compliance with regulatory requirements, safeguarding privacy, achieving superior performance, enhancing security measures, and ensuring cost-effectiveness. Enterprise data centers are designed specifically to align with an organization’s enterprise applications and processes. They can be located either on-premises or off-premises, depending on factors such as power availability, water accessibility, connectivity options, and security considerations.

Within an enterprise data center, an in-house IT department typically manages the servers, storage systems, and network equipment. However, the management of power equipment, cooling infrastructure, and security systems can be handled either internally or outsourced to a third party through a facility management contract.

In recent years, organizations have adopted a hybrid IT model combining cloud services with data center resources. This approach allows for a mix of on-premises infrastructure and utilization of colocation data centers. The shift towards the cloud is driven by scalability requirements and the need for faster time-to-market for new applications and services. Additionally, enterprise data centers often face challenges related to resource underutilization due to inadequate allocation, limited workload optimization, and a lack of visibility into resource usage.

Colocation Data Centers:

Colocation data centers, also referred to as multi-tenant data centers (MTDCs), offer a shared infrastructure for multiple organizations, typically 20 or more, to house their computing hardware, servers, and supporting facilities off-site. This is particularly beneficial for organizations that lack the physical space or IT resources to operate their own enterprise data centers, enabling them to allocate their personnel and financial resources to other important initiatives.

By leasing space within colocation data centers, organizations have the flexibility to easily scale their infrastructure up or down to meet their evolving needs. These facilities provide high availability, ample bandwidth capacity, and low-latency data access. To maintain these standards, colocation data centers regularly update and upgrade their hardware and technology.

Colocation data centers come in different sizes and cater to various data center requirements. Retail colocation data centers cater to smaller organizations, offering turnkey services within shared data halls. The data center operator manages day-to-day operations, power capacity, cooling, security, and telecommunication access, while customers are responsible for maintaining their equipment within designated areas.

Retail Colocation

Retail colocation services are suitable for smaller data center footprints and lower power capacity needs, typically ranging from 100 kilowatts (kW) to 1 megawatt (MW). Contracts for retail colocation generally last from 1 to 3 years, taking the form of leases or license agreements. Additionally, retail colocation providers offer value-added services such as remote hands assistance for equipment installation, troubleshooting, and maintenance, as well as additional network services such as connectivity to carriers or internet service providers (ISPs).

On the other hand, wholesale data centers are leased to a single customer who is responsible for building and managing the entire data center infrastructure. Unlike retail colocation providers, wholesale data center operators solely provide the space and power, without additional services like IT technicians or remote hands support. Wholesale data centers typically have fewer connectivity options to carriers and ISPs.

Wholesale Colocation

Wholesale data centers are designed to accommodate large enterprises, government agencies, and retail colocation providers. They offer white space rooms ranging from 10,000 to 20,000 square feet or larger. Leases for wholesale data centers are typically longer, ranging from 5 to 10 years, and are suitable for power capacity requirements ranging from 1 to 5 megawatts (MW).

Colocation data centers offer several advantages that make them a compelling option for businesses:

1. Reduced Capital Expenditure: Colocation allows businesses to leverage shared infrastructure, eliminating the need for significant upfront investments in real estate, equipment, and staffing. By renting space in a colocation facility, businesses can benefit from existing infrastructure and reduce initial costs.
2. Scalability: Colocation data centers offer the flexibility to scale infrastructure resources as needed. Businesses can start small and expand gradually without incurring major costs or disrupting operations, aligning their IT infrastructure with evolving needs.
3. Access to Specialised Expertise: Colocation providers have skilled professionals experienced in managing data center infrastructure, network operations, and security. By collocating, businesses can leverage this expertise without maintaining an in-house team, optimizing operations and addressing technical challenges effectively.
4. Improved Connectivity and Latency: With colocation, organizations have secure access to a growing selection of connectivity options, including multiple internet service providers (ISP), carriers and cloud providers—without building and managing the necessary connections and relationships. With a portfolio of geographically dispersed facilities, colocation providers also support latency requirements by allowing organizations to choose a facility near their headquarters.
5. Enhanced Security: Colocation facilities prioritize the security of infrastructure and data by implementing robust physical security measures. These include 24/7 surveillance, on-site security guards, card access terminals, biometric controls, turnstiles, and gate crash barriers. Such comprehensive security measures provide businesses with peace of mind, knowing that their critical assets are safeguarded within the colocation facility.

Comparison and Verdict

When considering the choice between enterprise data centers, retail colocation, and wholesale colocation, it is essential to assess your organization’s specific requirements and priorities. Here is a verdict based on the advantages and considerations of each option:

Enterprise data centers are ideal for organizations with highly customized IT infrastructure needs, compliance requirements, and the desire for complete control over their data center operations. Enterprise data centers offer the flexibility to tailor every aspect of the data center to specific requirements. They are suitable for organizations that have the resources and expertise to manage their infrastructure effectively.

Retail colocation data centers are a great choice for smaller organizations that lack the physical space or IT resources to operate their data centers. These facilities provide turn-key services, including infrastructure management, power capacity, cooling, security, and connectivity. Retail colocation offers flexibility, scalability, and the ability to benefit from shared infrastructure and specialized expertise without the need for extensive upfront investments.

Wholesale colocation data centers are designed for larger enterprises, government agencies, and retail colocation providers. They offer larger white space rooms and longer lease terms, catering to organizations with higher power capacity requirements. Wholesale colocation is suitable for organizations that have the resources and expertise to build and manage their entire data center infrastructure, as the operator mainly provides space and power without additional services.

Data Centers and Business Continuity Disaster Recovery Planning: Safeguarding Your Business Against Disruptions

Business Continuity and Disaster Recovery, or BCDR, refers to the processes and procedures that organizations implement to ensure the continuity of critical business functions during and after a disaster or disruptive event. The primary goal of BCDR planning is to reduce the impact of disruptions, maintain business continuity, and quickly return to normal operations.

BCDR is crucial because downtime in a data center can have significant negative consequences for companies. It can lead to lost revenue, decreased productivity, missed opportunities, and damage to customer satisfaction and reputation. According to an IDC survey, for Fortune 1,000 companies, downtime can cost as much as $1 million per hour. With that said, when it comes to global giants like Amazon, whose entire business model relies on uptime, estimated costs of downtime are around $13.22 million per hour.

Data centers with a Disaster Recovery Center provide a contingency and real-time response during emergencies. These centers are equipped with the necessary infrastructure to quickly connect to enterprises’ disaster recovery systems, enabling them to resume operations promptly and respond instantly. Businesses should also seek data centers with 24/7 customer support and that regularly conduct disaster drill tests for enhanced safety and reliability.

In addition to mitigating operational risks, BCDR is often mandated by regulatory and compliance frameworks like HIPAA, PCI DSS, and SOX. Failure to comply with these standards can result in legal and financial penalties. Therefore, having robust BCDR plans and processes in place is not only essential for maintaining business operations but also for meeting regulatory requirements and avoiding potential consequences.

On March 10 2021, a fire broke out at OVHcloud’s SBG2 data center in Strasbourg, France. This unexpected incident resulted in the complete destruction of several buildings within the facility, surprising industry professionals worldwide. The service provider’s founder Octave Klaba, candidly shared updates about the extensive damage on Twitter, urging clients to activate their Disaster Recovery Plans. Unfortunately, many clients discovered that their DR plans did not consider the possibility of a fire affecting multiple buildings within the same facility. Those who relied on data backups located on different servers within the same campus were left without backups.

For complete assurance of uninterrupted operations, it’s highly recommend implementing true geodiversity. This involves locating your production systems and backup systems in separate regions to safeguard against prolonged disruptions caused by power outages or facility losses in a particular state or facility.

Expert On-Site Assistance: ‘Remote Hands’ as a Service

Remote hands are a service provided by data center operators that offer on-site assistance and support for various tasks related to the management and maintenance of IT infrastructure. These tasks can include equipment installation, troubleshooting, maintenance, physical inspections, and basic IT support. Remote hands services are typically activated when businesses encounter technical issues or require physical assistance with their equipment located in the data center.

Businesses require remote hands services for several reasons. Firstly, it may not be feasible or cost-effective for businesses to have their personnel available at the data center facility at all times. Remote hands services provide businesses with a dedicated team of experts who can promptly address issues and perform tasks on their behalf. This saves businesses time, resources, and the need for physical presence. Furthermore, remote hands services offer convenience and flexibility to businesses. They can access skilled technicians 24/7, ensuring swift response and resolution of any issues that may arise. This allows businesses to maintain the smooth operation of their IT infrastructure without delays or disruptions.

The benefits of remote hands services for businesses are significant. It provides access to specialized expertise, ensuring that tasks are performed efficiently and correctly by experienced professionals. Businesses can rely on the data center operator’s resources, infrastructure, and technical knowledge to overcome challenges and maintain optimal performance. Remote hands services also reduce downtime, enhance IT operations’ reliability, and free internal resources to focus on core business activities.

Key Factors to Consider When Choosing a Data Center

• Location: Each country offers unique attributes that make it desirable for businesses to choose data centers in those locations. Singapore is known for its geographical stability and provides a safe choice for establishing a data center. Hong Kong is popular among businesses looking to leverage its connectivity to China and benefit from its strong financial and telecommunications sectors. Japan stands out for its advanced technology infrastructure, stable power supply, and robust disaster preparedness. Data centers in China offer businesses the opportunity to establish a presence in one of the world’s largest economies and cater to the specific demands of the Chinese market.
• Network Ecosystem: The network ecosystem within a shared data center space is a valuable advantage of colocation. It allows businesses to establish connections with various entities such as partners, distributors, and even competitors. This interconnectivity fosters collaboration, facilitates data exchange, and opens up opportunities for business growth. When choosing a data center, it is crucial to evaluate the available connectivity options. Additionally, consider if the data center is already connected to a large ecosystem of customers, as this indicates a robust and thriving network environment that can enhance your connectivity options and facilitate valuable partnerships.
• Support, Services, and Monitoring: Having comprehensive support services and robust monitoring capabilities is critical for keeping your data center running smoothly. A Network Operations Center (NOC) staffed by experienced professionals can monitor your infrastructure proactively, identify potential issues, and provide timely resolutions. Incident response, regular maintenance, and updates are critical components for ensuring your data center’s ongoing reliability and performance. Consider the level of support and monitoring provided by data center providers when evaluating them, ensuring that it meets your specific needs.
• Power Usage Effectiveness: PUE is a metric that measures a data center’s energy efficiency. It calculates the amount of energy used by the data center to power IT equipment, cooling systems, and other operational functions. A lower PUE indicates greater energy efficiency, which leads to lower energy consumption and cost savings. Inquire about a data center’s PUE rating when choosing one to ensure that they prioritize energy-efficient practices. You can contribute to sustainability efforts and potentially lower your operational expenses by selecting a data center with a low PUE.
• Service Level Agreements: Service Level Agreements (SLAs) are agreements that outline the performance and reliability commitments between you and the data center provider. These agreements specify the expected uptime, availability, and response times for your applications and services. When assessing data center providers, it is important to carefully review their SLAs to ensure they align with your organization’s needs and objectives.

Data Centers at China Telecom Asia Pacific

Embarking on a journey of digital transformation requires a reliable and innovative data center infrastructure, and our Singapore region hosts an impressive array of facilities to meet those demands.

From the cutting-edge Singapore Woodlands Data Center and the strategically located Singapore Tai Seng Data Center to the technological hub that is the Japan Tokyo Data Centre and our expanding presence in Indonesia, we offer a comprehensive network of data centers. These facilities provide access to a diverse range of global and regional carriers and ISP nodes, ensuring connectivity excellence.

Our services encompass dark fiber, wave channel services, Ethernet, MPLS, VPLS, internet transition, internet peering connection, CDN, co-location, and more. What sets us apart is our ability to tailor solutions to the unique needs of our clients. Whether it’s public cloud, private cloud, hybrid cloud, or ICT solutions, we take pride in delivering customised strategies that empower businesses to overcome communication challenges across the dynamic landscape of the Asia-Pacific region.

Get in Touch with China Telecom: Unlock the Power of Colocation Data Centers

China Telecom understands the importance of seamless connectivity, unmatched performance, and tailored solutions when it comes to data centers. With their comprehensive range of services, they stand ready to help businesses of all sizes embrace the power of advanced data management.

As an end-to-end partner, China Telecom simplifies workflows by delivering all required IT solutions under one roof, streamlining coordination efforts. China Telecom holds a commanding presence in the telecommunications sector, reigning as a global and domestic leader with an expansive network and substantial customer reach. Their global network resources ensure that operations maintain seamless connectivity across different locations, allowing for smooth processes and enhanced productivity. With China Telecom’s professional project management team, businesses can rely on their experience and resources for comprehensive support throughout their digitalization journey.

Each business has unique needs, which is why China Telecom’s solutions are fully customizable. China Telecom is dedicated to meeting businesses’ specific requirements by providing the most relevant and advanced solutions in the market. Security is a top priority of theirs. With managed security services, China Telecom offers heightened security practices tailored to meet your organization’s specific security needs. You can have peace of mind knowing that your business is always protected.

Experience the unparalleled benefits of colocation data centers with China Telecom and revolutionize your IT infrastructure. Take the next step towards optimizing your data management by getting in touch with China Telecom today.