Total Power
10 MW
MACALLEN AI COLOCATION
High-Density AI Colocation:
MacAllen Site
Redundant Web Services (RWS) delivers dedicated, high-performance compute infrastructure for demanding AI and deep learning workloads. Positioned for low-latency traffic patterns and hybrid delivery, our MacAllen site gives operators an edge-ready South Texas footprint with flexible capacity.
10 MW
Site capacity
90d
Deploy time ARO
75%
vs hyperscaler cost
Accelerating the NVIDIA AI Infrastructure Frontier
HIGHLIGHTS
Facility & Architecture Highlights
-
Deployment Velocity
90 to 120-day lead time ARO for custom colocation build-outs.
-
Network Readiness
Sites are ready to populate with lite & dark fiber available immediately.
-
Next-Gen Hardware Support
Fully equipped to host the complete NVIDIA portfolio, including A100, H100, H200, B200, B300, and GB300 systems.
-
Container-Scale Deployment
Built on highly efficient 120 kW modules that are fully stackable. Our architecture is designed to seamlessly handle hyperscale rollouts of 50 to 200 containers per site.
-
Advanced Cooling
Each container module is adaptable and equipped to support both air and water cooling configurations.
-
Geographic Advantage
MacAllen's South Texas position supports latency-sensitive services, hybrid delivery models, and regional resiliency for traffic that needs to stay close to the edge.
MACALLEN BENEFITS
The MacAllen Advantage
MacAllen combines low-latency regional reach, hybrid-friendly traffic patterns, and modular infrastructure built for responsive GPU and edge deployments.
A100 Capacity
6,000
H200 Capacity
5,500
B200 Capacity
5,000
SITE SPECIFICATIONS
Other Site Capacities & GPU Densities
Our infrastructure allows you to match infrastructure to workload without compromising performance or efficiency.
| Site Location | Total Power | A100 Capacity | H200 Capacity | B200 Capacity | H100, B300, GB300 |
|---|---|---|---|---|---|
| Odessa, TX | 24 MW | 14,400 | 13,200 | 12,000 | Supported |
| George West, TX | 13 MW | 7,800 | 7,150 | 6,500 | Supported |
| Wenatchee #1, WA | 6 MW | 3,600 | 3,300 | 3,000 | Supported |
GPU capacities are based on standard per-MW densities inside our 120 kW modules: A100 (600/MW), H200 (550/MW), and B200 (500/MW).
WHY RWS
The Advantage of Partnering With RWS
Enterprise Reliability without the Drag
Skip the 12 to 18-month industry standard hyperscaler wait times and deploy faster.
Predictable Economics
Transparent, reserved pricing resulting in 40–75% cost savings compared to hyperscaler on-demand pricing.
Architectural Freedom
We support tenant-supplied hardware, hybrid colocation, or RWS-provided hardware with zero hyperscaler vendor lock-in.
Contact Us to Reserve Capacity
To discuss site tours, technical validation, or to reserve your colocation deployment, please contact: