News & Insights
Optimizing space utilization and flexibility in colocation data centers
Colocated data centers, which lease space to multiple tenants, play a critical role in the data center ecosystem. Yet, the growing demands for capacity pose significant challenges. Read more to find out why and what opportunities are available for optimizing utilization while maintaining flexibility.

A colocation data center (colo), is a facility where businesses can rent/lease space to house their IT infrastructure, such as servers and networking equipment. The colo provides the physical environment, including power, cooling, bandwidth, and security, while the businesses retain control over their equipment and software.
Demand for these centers is increasing not only for their traditional clients who need smaller scale solutions verses building their own data center. Owners of colos are feeling the need to maximize space utilization while maintaining operational flexibility to support growth.
While phased development helps colos expand incrementally and manage first costs, it can result in underutilized space and resource inefficiencies in the short term. This challenge is further complicated by the evolving nature of client workloads, which require diverse power, cooling, and networking configurations that must remain adaptable for future needs. The pressure to deploy quickly, particularly in competitive markets, further amplifies this pressure. Compliance with regional regulations and sustainability goals introduces another layer of complexity, requiring colos to strike a balance between performance, cost-efficiency, and environmental responsibility.
Salas O’Brien works with colocation data center clients to resolve these issues. Here are some key strategies for resolving space and engineering challenges for colo data centers:
Core strategies for efficient incremental growth
Data centers can be very large buildings, averaging between 85,000 to 250,000 square feet with imposing heights of 35–40 feet for single-story designs. Despite their size, only a small fraction, 5 to 15% of the floor space is allocated for human use, primarily for security and minimal technical maintenance/operational personnel. The rest is dedicated to the computer rooms with racks of servers, and the mechanical, electrical and plumbing (MEP) equipment necessary to operate the servers and protect them from ever losing power. In addition to the building’s interior, there are large exterior utility yards for equipment such as generators, chillers, or cooling towers.
Data centers have long been designed for building out data halls and their support spaces by ‘phase’ allowing the owner to construct the building, administration, and first phase while leaving remaining portions of the building to be built out at a later date. This allows the owner to manage their first costs and build future phases as the clients are secured. Another effective growth strategy is the use of modular designs, which enable data centers to expand capacity incrementally while maintaining efficient use of resources. Modular construction—whether in the form of pre-engineered/containerized components such as electrical equipment or even smaller fully containerized data centers that are connected together—offers speed and flexibility, reducing deployment times and minimizing disruption to existing operations. These solutions also allow for standardized configurations that simplify maintenance and reduce costs, making them an effective option for colos looking to manage growth dynamically.
In addition to modularity, space partitioning and multi-purpose infrastructure play a vital role in adaptability. By designing spaces that can be reconfigured to meet diverse power, cooling, and networking needs, colos can better accommodate the unpredictable workloads of their clients. Multi-purpose infrastructure, such as shared cooling systems and flexible electrical distribution, further enhances this adaptability, allowing data centers to serve a broader range of applications without requiring extensive redesigns. Dividing data halls into distinct zones for staggered buildouts facilitates efficient use of resources while maintaining flexibility for future expansion. These zones are often separated by security levels, with higher security protocols applied to data halls than to administrative areas.
In addition to managing floor space, it is important to manage the vertical space. Server racks are getting more dense as the computers in them get smaller/flatter. Racks themselves average 7 to 7’-6” tall but house more computers. With more computers comes more power connections and more infrastructure connections such as fiberoptic cabling. These support systems get stacked above the server racks often creating an overall height of 12 to 14 feet. Taller rack and support systems requires taller ceilings and robust airflow management. Maximizing vertical efficiency is critical for operator access to the systems as well as efficient heat removal. Salas O’Brien recently worked with a coloprovidor in Washington to leverage hot air plenum systems above the ceilings to help direct and extract heat efficiently, leveraging the building’s cubic volume.
Adaptability for evolving workloads
While the evolution of server technology has enabled tolerances for higher ambient temperatures (from 65°F to 85°F or higher) which reduces cooling costs, temperature control remains one of the biggest challenges for data centers. Dynamic cooling systems can provide precise control over airflow and temperature, allowing data centers to optimize energy use based on real-time needs. Leveraging technologies like variable refrigerant flow (VRF), hot-aisle containment, and passive economizer cycles that use external air when conditions permit can drive greater efficiencies. The overarching goal is to design systems that move cooling outputs as close to the heat source as possible while minimizing energy consumption.
On the networking side, software-defined networking (SDN) enables colos to respond to shifting client demands with agility. By virtualizing and automating network management, SDN allows for rapid reconfiguration of network paths, bandwidth allocation, and security protocols without requiring physical changes to infrastructure.
Accelerated deployment for speed-to-market
Turning a colocation data center from a concept into a revenue-generating asset requires a sharp focus on speed-to-market. Meeting existing demand hinges on how quickly the facility becomes operational.
While building information modeling (BIM) is a staple for most architects and engineers, leveraging BIM models for detailed phasing and sequencing along with construction estimation and procurement can accelerate timelines while reducing errors and enhancing coordination during construction phases.
Virtual design and construction services allow stakeholders to anticipate and address potential conflicts, such as equipment spacing or routing issues before they occur in the field. Salas O’Brien is currently providing MEP and controls engineering, building information modeling, and virtual design and construction services for a cutting-edge 400 MW hyperscale data center with a goal of speed-to-market.
In addition, streamlined commissioning processes ensure that data centers achieve operational readiness quickly without compromising reliability. Commissioning integrates testing and verification of all critical systems, from power and cooling to networking infrastructure. By adopting ongoing commissioning methodologies, colos can validate systems as they are installed, rather than waiting until the end of the construction process. This proactive approach not only saves time but also reduces the likelihood of costly delays or failures during initial operations.
Balancing sustainability with operational goals
Colocation data centers require flexibility in energy management. While net-zero operations remain an aspirational goal for most due to their immense power demands, incremental advancements in energy efficiency are becoming increasingly impactful. One area of focus is renewable energy integration in which colos are leveraging on-site solar installations, wind power agreements, or energy storage systems to reduce reliance on grid electricity. In some instances, on site generation can store energy and provide it back to the utility grid for use in other parts of the jurisdiction. These renewable energy strategies not only decrease carbon footprints but also provide cost stability in regions with volatile energy markets.
By prioritizing passive cooling methods, such as airside economization or advanced evaporative cooling, colos can significantly lower their energy use while maintaining reliable performance. Combining these approaches with recyclable construction materials and energy-efficient designs allows data centers to achieve certifications like LEED without excessive investment.
Physical security needs of co-located data centers
Data centers require stringent security measures, from multi-layered access controls to fortified building facades and perimeters. Architectural designs must comply with local jurisdictional requirements, which can include everything from gate height regulations to façade aesthetics. These requirements can impede security measures which can lead to additional precautions such as exterior cameras monitoring the area. As colocation data centers serve diverse tenants, physical security requires a layered approach which is adaptable but secure as the needs within the center change.
Partitioning facilities by access levels minimizes risk while maintaining efficiency. Robust access controls, such as biometric scanners, RFID-enabled locks, and multi-factor authentication, ensure that only authorized personnel can enter sensitive areas like data halls and control rooms.
Advanced video surveillance systems equipped with AI-powered analytics can detect and respond to unusual activity in real-time. These systems provide comprehensive monitoring across entry points, shared spaces, and critical infrastructure zones.
Enhanced physical barriers, such as fencing, bollards, and intrusion detection systems, provide additional layers of protection to prevent unauthorized access before individuals reach the facility itself. For colos in urban areas, discrete but effective perimeter design aligns with both security and aesthetic goals.
Most importantly, integration of redundant systems ensures that a single point of failure does not compromise overall security. This includes backup power for security systems, mirrored monitoring stations, and regular system testing.
How Salas O’Brien can help
Salas O’Brien can provide tangible solutions for space utilization, energy efficiency, and scalability while helping data centers maintain flexibility. Our deep bench of multidisciplinary architects, engineers, sustainability consultants, physical security designers, and commissioning agents have decades of experience in serving data center clients. Whether hyperscale, enterprise, or colocation data centers, we can help meet your facility needs. Reach out to our contributors below.
For media inquiries on this article, reach out to Stacy Lake, Director of Corporate Communications.

Brenda Ross, NCARB, AIA
As head critical environment architect, Brenda Ross excels in delivering innovative design solutions that meticulously blend the art of architecture with the reliability imperative for mission-critical projects. With an unwavering attention to detail, she skillfully integrates building design and engineering, resulting in facilities that are elegant, secure, reliable, and scalable. Distinguished as a leading expert, Brenda spearheads multidisciplinary design teams, offering project leadership and deep technical expertise to a wide array of critical environments. She serves as Principal Architect, Director of Critical Environments at Salas O’Brien. Contact her at [email protected].