Data center decommissioning plays a crucial role in a range of broader services, including data center migration, consolidation, and upgrades, all of which aim to modernize and update the hardware and infrastructure of companies. This decommissioning step is essential for achieving cost savings, enhancing efficiency, improving security, and gaining environmental benefits in data centers and their key components, such as servers. Its significance is particularly notable in the context of digital transformation initiatives and IT asset disposition (ITAD) projects.

Data center decommissioning is the process of systematically shutting down and dismantling the operations and equipment of a computer facility. This includes the removal of hardware, proper disposal or repurposing of equipment, and ensuring data security and environmental compliance in the process.

Dgtl Infra delves into the intricacies and best practices of effectively shutting down data centers. From exploring the reasons behind data center decommissioning to discussing the detailed steps involved in the process such as data migration, physical dismantling, and secure data destruction, this article serves as a thorough guide. Continue reading to gain insights on selecting the right decommissioning services and companies, as well as additional focus areas like server and application decommissioning, complete with case studies and practical checklists.

What is Data Center Decommissioning?

Data center decommissioning is the process of systematically shutting down and dismantling the operations and equipment of a data center. This process includes removing and disposing of hardware like servers, ensuring data is either transferred or securely erased, and addressing physical infrastructure elements like cabling, power equipment, and cooling systems. The goal of decommissioning a data center is to minimize risks, such as data breaches or environmental impacts, while efficiently reallocating or disposing of resources.

Data Center Interior showing Servers Ladders Tools Raised Flooring Decommissioning

Reasons for Decommissioning a Data Center

Organizations may decide to decommission a data center for various reasons:

  1. Outdated Technology: Technology rapidly evolves, rendering data center components obsolete. Decommissioning allows for the replacement of outdated servers, storage systems, networking equipment, Uninterruptible Power Supply (UPS) systems, Power Distribution Units (PDUs), backup generators, air conditioning (CRAC and CRAH) units, ventilation systems, and cooling apparatus. This upgrade to newer technologies enhances a data center’s performance and capabilities
  2. Cost Reduction: Operating and maintaining a data center incurs significant expenses, including power, cooling, maintenance, and staffing. Decommissioning, particularly of underutilized or inefficient facilities, can substantially reduce these costs. Underutilization may manifest in low utilization of IT resources (like CPU and storage systems) or power, referring to the proportion of electrical power consumed by the data center that is actually used by IT equipment
  3. Consolidation and Efficiency: Companies often decommission data centers as part of a strategy to consolidate their IT infrastructure. This data center consolidation increases efficiency through better resource utilization, reduced complexity, and a shift toward scalable solutions such as colocation or cloud computing. Organizations may even fully transition away from on-premises data centers to these services, reducing overhead. Such changes often accompany major corporate events like mergers, acquisitions, restructurings, or downsizings
  4. Security and Compliance: To align with evolving data security and privacy laws (e.g., Europe’s GDPR), companies may decommission data centers. This move helps strengthen compliance and risk management by minimizing vulnerabilities and the attack surface associated with outdated or redundant technology infrastructure
  5. Environmental Concerns: Older data centers typically have lower energy efficiency and a larger carbon footprint. Decommissioning them in favor of modern, energy-efficient technologies can improve Power Usage Effectiveness (PUE) – a metric for data center energy efficiency – thereby reducing an organization’s environmental impact

Data Center Decommissioning Process

Decommissioning a data center is a complex and important process that involves several distinct steps.

Digital Information Represents Process with Several Distinct Steps and Stages

Here is an outline of the typical data center decommissioning process:

1. Planning and Asset Inventory

  • Develop a comprehensive plan to outline the project’s scope, establish timelines considering the potential impact on business and resource availability, formulate a budget, and evaluate risks
  • Assess the implications for business operations and prepare for data migration if needed
  • Perform an inventory of all physical assets and equipment scheduled for decommissioning
  • Decide whether to use external contractors or internal staff, like system administrators or engineers, for the physical decommissioning of the data center
  • Plan the termination of any vendor maintenance contracts and licenses for servers or software not being migrated
  • Utilize Data Center Infrastructure Management (DCIM) software for maintaining an accurate, real-time inventory of physical data center assets. This software also helps in scheduling tasks and allocating resources for the decommissioning project

2. Data Migration and Backup

Before proceeding with decommissioning, ensure the secure backup or migration of all critical data to a new location, which could be another data center or cloud storage. Next, conduct thorough tests and simulations on all backups to ensure their reliability. Crucially, confirm the integrity of the data following its transfer to guarantee no loss or corruption has occurred.

3. Physical Decommissioning

This stage involves shutting down and removing servers, storage systems, switches, routers, and other network equipment from their racks. It includes disconnecting and uninstalling power supplies, cooling systems, racks, cabinets, rail kits, power cords, keyboards, cabling, and raised floors. Additionally, equipment must be removed from access control lists (ACLs), subnets, and firewalls.

Power to all equipment scheduled for decommissioning should be cut off, and virtual machines (VMs) being retired must be properly decommissioned. It is important to reconcile the inventory count of the equipment, documenting any discrepancies against the original asset inventory list.

4. Asset Reuse and Disposal

The next step is identifying assets suitable for reuse or repurposing. These assets should be packed, shrink-wrapped onto pallets, prepared for shipment, tracked, and relocated to a new data center. Due to the size and weight of data center hardware, machinery such as cranes, forklifts, hoists, tip guards, and pallet jacks are necessary for transporting the hardware.

Assets Reused and Repurposed with the help of a Crane Loading Equipment onto Truck

To ensure a secure chain of custody, the transportation vehicles must be locked and sealed to safeguard the hardware from theft or damage during transit.

Equipment that is no longer required should be responsibly disposed of. This may involve selling, recycling, or destroying (via shredding or degaussing) the equipment. While equipment that is in good condition and relatively new can often be resold, equipment that is obsolete or damaged should be recycled or disposed of properly.

5. Data Destruction and Security

The foremost priority is securing the information held within the data center. All data on decommissioned equipment must be erased or destroyed using established industry practices. This crucial step prevents unauthorized access to sensitive information, such as personally identifiable information (PII) and confidential business data.

6. Facility Cleanup

After equipment removal, the facility must undergo thorough cleaning. This involves the removal of racks, cabling, and other infrastructure items. For leased data centers, the lease agreement specifies the required condition for returning the facility to the landlord, including any necessary repairs or maintenance to restore the property to its original state.

7. Final Assessment and Documentation

Conduct a final assessment to ensure that all data center decommissioning steps have been effectively completed. Documentation of the entire decommissioning process should be completed for future reference. It supports accurate accounting for asset write-offs and ensures compliance with environmental and regulatory standards.

This phase also includes asset reporting, which records the final status, resale value of hardware, and key dates related to the resale, recycling, or redeployment of assets. Recording this data provides a comprehensive audit trail. Documentation is similarly vital for confirming the termination of maintenance contracts and acquiring certificates of data erasure or hardware destruction.

Data Center Decommissioning Services

Decommissioning a data center involves a range of services to safely and efficiently dismantle, remove, and dispose of or repurpose the data center’s infrastructure.

Provider Delivering Data Center Decommissioning Services

Here are common data center decommissioning services:

1. Asset Inventory and Auditing

This crucial service involves compiling a detailed inventory of all assets in the data center. It includes cataloging a range of elements, such as hardware, software, operating systems, and applications to determine what needs decommissioning, recycling, or repurposing. The inventory typically includes specifics like the device name, serial number, IP address, vendor, model, as well as processor, memory, and storage capacities.

Thorough documentation of inventory is essential for replacing obsolete equipment and replicating necessary resources to meet workload demands. The process involves conducting both a physical audit of hardware and a software discovery audit. These findings are then compared with the existing configuration management database (CMDB). This comparison enables organizations to cross-check and update their records of all hardware and software assets, including the documentation of all asset dependencies.

Ultimately, a complete and accurate inventory list allows for the creation of a data center environment map. This map identifies dependencies for various elements such as servers, storage systems, networking equipment, cabling, power and cooling systems, software, and virtual machines (VMs).

2. Data Sanitization and Destruction

A crucial aspect of decommissioning is ensuring data security. Data sanitization is the process of securely wiping data from storage devices, either through software or by physically destroying the devices. This includes servers, hard drives, and other forms of storage. The goal is to render the data unrecoverable, thereby preventing potential data breaches.

Typically, companies choose on-site destruction of digital data during the decommissioning process. This can be achieved through physical destruction methods or data erasure, depending on the type of storage device. A key reference for these practices is the NIST Special Publication 800-88, “Guidelines for Media Sanitization,” which provides comprehensive standards for securely erasing and disposing of various digital media, including IT equipment in data centers.

Additionally, decommissioning vendors issue Certificates of Data Destruction (CDD) to confirm that the data destruction process complies with both corporate policies and regulatory standards, such as the General Data Protection Regulation (GDPR) in Europe.

3. Physical Dismantling and Removal

This step involves the methodical dismantling of data center components such as racks, servers, power and cooling systems, and other related infrastructure. It also includes their subsequent removal from the data center facility.

Given the specialized nature of data center components, such as Uninterruptible Power Supply (UPS) systems, heat exchangers, and backup generators, specialized contractors are often employed. These professionals ensure that the power and cooling systems, which are unique to data centers, are decommissioned and dismantled safely and efficiently.

Shutting Down and Dismantling Large Empty Server Farm in White Space

In certain cases, demolition may be the preferred approach. This is particularly relevant for data centers or specific infrastructure elements that are either too outdated for reuse or resale, or so extensively damaged that dismantling or removal would be impractical or hazardous.

Infrastructure commonly subjected to demolition includes custom-built server racks and cooling systems, as well as components that are integrated into the building’s structure, like cooling ducts and raised floor systems.

4. Asset Resale and Recycling

Proper disposal of electronic waste (e-waste) is crucial for environmental compliance. This service, often referred to as IT asset disposition (ITAD), involves reselling and recycling components such as servers, storage devices, network equipment, uninterruptible power supplies (UPS), and memory modules.

If the decommissioned equipment is not needed internally, it can be resold on behalf of the data center. This process includes evaluating the market value of assets and finding buyers or new uses for them, which helps offset the costs of decommissioning, data destruction, relocation, and reporting.

Asset disposal must adhere to legal and environmental guidelines, as well as industry standards, including the R2 (Responsible Recycling) certification and the ISO 14001 standard. ITAD service providers, typically accredited for the safe and environmentally responsible disposal and recycling of IT equipment, can provide certificates of recycling.

Green Recycling Bin Filled with Electronic Waste also known as E-Waste

The recycling process involves separating scrap materials like steel, aluminum, plastic, copper, glass, and precious metals (for example, gold and silver). This separation allows these commodities to be reused in the manufacturing of next-generation products.

Data Center Decommissioning Companies

Data center decommissioning companies, typically operating as a branch of IT asset disposition (ITAD) providers, specialize in securely erasing data from servers, hard drives, and other storage components. Their services extend to dismantling the data center equipment and then selling the components under a revenue-sharing model. In this arrangement, the data center decommissioning company earns a percentage of the sales proceeds while not having to own the inventory they are selling.

Technician Handles Computer Hardware for the purposes of IT Asset Disposition ITAD

Examples of data center decommissioning companies include:

  1. Iron Mountain: Offers decommissioning services to hyperscale businesses, including cloud service providers and large enterprises. It owns ITAD providers ITRenew and Regency Technologies
  2. Sims Lifecycle Services (SLS): SLS provides both ITAD and decommissioning services. They focus on the reuse, redeployment, and recycling of IT and data center equipment
  3. Wisetek: As a global ITAD and technology service company, Wisetek specializes in secure data destruction, reuse, and recycling solutions. They are committed to environmental sustainability and adhere to circular economy principles
  4. TES amm: Specializing in sustainable technology lifecycle management, TES amm offers comprehensive solutions for transforming and repurposing IT devices in an environmentally responsible manner
  5. Apto Solutions: As an ITAD service provider, Apto focuses on reducing data, environmental, and financial risks for clients across various sectors. Their emphasis is on sustainable practices

Additionally, enterprise IT solutions providers like Hewlett Packard Enterprise (HPE) and Dell Technologies offer data center decommission services through their ITAD and asset recovery divisions.

Case Study – An Example of Data Center Decommissioning

Iron Mountain, a leader in this field, recently shared a case study illustrating their data center decommissioning services for a major U.S. bank. This bank, having acquired a competitor, faced the challenge of redundant and costly data centers.

Iron Mountain’s services included not only data center decommissioning but also data erasure, on-site media destruction, and asset tagging.

A Case Study on Iron Mountain of Data Center Decommissioning for a U.S. Bank

The solutions provided by Iron Mountain were customized to align with the bank’s governance and risk mitigation policies. This involved creating a tailored decommissioning workflow process and deploying on-site technicians to effectively dismantle the bank’s data centers and assets.

Key achievements by Iron Mountain, all in compliance with the bank’s stringent standards and timeline, were:

  • 13,000 hard drives securely removed and shredded on-site
  • 1,300 servers de-racked and processed across 15 data centers

What is Server Decommissioning?

Server decommissioning is the process of safely and systematically removing a server from active service, typically when it is no longer needed, has become obsolete, or is malfunctioning. This involves ensuring that all sensitive data is securely erased or transferred, and the server’s software and hardware components are properly disposed of or repurposed. The process also includes documenting the decommissioning for compliance and auditing purposes, and updating the organization’s inventory and network configurations to reflect the change.

Server Decommissioning Decommission Process Equipment Inside Data Center Facility

Major cloud service providers (CSPs) like Microsoft and Google typically refresh their server fleets every 5 to 6 years, part of which involves decommissioning outdated technology in their data centers. Servers may also be decommissioned for various other reasons, including low utilization rates, lack of use in ‘hot standby’ scenarios, software incompatibilities, or for improving cost efficiency.

How to Decommission a Server – Process and Checklist

Decommissioning a server is a critical process in IT management and data centers, involving several steps to ensure that the server is safely and securely taken out of service. The entire process to decommission a server typically takes two to six weeks, depending on several factors, such as the complexity of the server and the data stored on it. Here is a general checklist for decommissioning a server:

1. Backup Data

Before starting the server decommissioning process, it is crucial to back up all important data stored on the server. This ensures that no critical information is lost. Use reliable backup tools and verify the integrity of the backup. Run tests to ensure that backup process and disaster recovery are functioning.

2. Notify Stakeholders

Inform all relevant stakeholders about the planned server decommissioning. This includes IT staff, users who rely on the server, and any external parties that might be affected. Provide details about the timeline and any expected disruptions.

3. Shut Down Services and Applications

Gradually shut down the services and applications running on the server. Concurrently, terminate the associated software licenses and vendor maintenance contracts specific to this server. Ensure that this process is managed in a controlled manner to prevent data loss or corruption. If the server is part of a cluster or network, adjust the configurations to accommodate the change.

4. Physically Disconnect the Server

Once all services are stopped, physically disconnect the server from the network. This includes unplugging Ethernet cables, storage connections, and power supplies. Next, remove all subnets, access control lists (ACLs), and firewalls associated with the server. Once powered down, remove the server from its rack.

Completing these steps is crucial for preventing the formation of security gaps in the network, which could otherwise become vulnerabilities that cyber attackers might exploit.

5. Secure Data Erasure

Perform a hard data wipe to erase all stored data on the server, ensuring an irreversible deletion process. This step is crucial for maintaining data security and privacy, particularly if the server contained sensitive information. Utilize certified data destruction software for this data erasure, and ensure it is done while the server is still in its rack.

If data wiping software is not used, physically destroy the server’s storage systems through shredding, degaussing, or crushing. This approach is ultimately the most effective in safeguarding sensitive data on a hard drive.

6. Audit and Documentation

Conduct an audit to ensure that all steps have been completed successfully. Document the entire process, including data backup, shutdown procedures, and data erasure confirmation. This documentation, including any certificate of erasure or destruction, is essential for compliance and future reference.

Additionally, it is important to update and consult certain organizational documents. This includes asset management records, which help maintain accurate inventory information, and change management documents, which are vital for ensuring that proper processes are adhered to throughout the operation.

7. Disposal or Repurposing

Decide on the final disposition of the server. If the server is still functional, it could be repurposed or sold. If it is no longer usable, place the server on a pallet and remove it from the data center facility. Follow environmentally responsible methods to dispose of the hardware, complying with electronic waste (e-waste) regulations.

What is Application Decommissioning?

Application decommissioning is the process of formally ending the life of a software application. This involves removing the application from active service, and typically includes the safe migration or archival of its data. The goal is to ensure that all associated resources are reallocated, data is secured or disposed of appropriately, and any potential risks or dependencies are managed efficiently.

Mary Zhang covers Data Centers for Dgtl Infra, including Equinix (NASDAQ: EQIX), Digital Realty (NYSE: DLR), CyrusOne, CoreSite Realty, QTS Realty, Switch Inc, Iron Mountain (NYSE: IRM), Cyxtera (NASDAQ: CYXT), and many more. Within Data Centers, Mary focuses on the sub-sectors of hyperscale, enterprise / colocation, cloud service providers, and edge computing. Mary has over 5 years of experience in research and writing for Data Centers.

LEAVE A REPLY

Please enter your comment!
Please enter your name here