Skip to main content Skip to main content

RES AI GPU Servers

Supercomputing for the Edge  

Leveraging the latest processing, co-processing, and accelerator technologies, RES AI brings High performance computing (HPC) to defense and IOT applications at the Edge. 


Engineered to Tackle Challenging Workloads Anywhere

Designed—and proven—to handle massive Artificial Intelligence (AI), virtualization, big data an alytics, Signal Intelligence (SIGINT), cryptography, image processing, Electronic Warfare (EW) and sensor fusion workloads anywhere, RES AI servers integrate the latest NVIDIA® Tesla® GPUs and Intel® Xeon® Scalable processors in environmentally resilient, rugged form-factors. Multiple expansion slots support specialized capabilities while shallow, densely packed chassis maximize performance in a small footprint. 


Maximize Throughput with Parallel Computing

Powered by multiple NVIDIA® Volta, Pascal™ and Turing™ architecture GPUs, RES AI servers harness parallel processing to maximize throughput, boost productivity and push the boundaries of compute-heavy applications. Offering the performance of up to 100 CPUs in a single GPU, NVIDIA’s Tesla® V100 data center GPU optimizes single-, double-, and mixed-precision workloads with 5,120 CUDA and 640 Tensor cores. With the same core-count, the GV100 GPU accelerats ray tracing, AI, simulation, and VR enhanced workflows. A single RES AI 4U server supports eight Tesla® V100 or GV100 data center GPUs. 


RES AI XR6 3U, 8 drive 24 Inch

  • Up to four NVIDIA Tesla V100 or GV100 GPUs with 7 PCIe 3.0 slots
  • Two Intel Xeon Scalable CPUs and 4TB memory
  • 240TB storage with eight 2.5" or 3.5" drives
  • MIL-STD: 810G, redundant power


RES AI XR6 3U, 16 drive 21 inch

  • Up to four NVIDIA Tesla V100 or GV100 GPUs with 6 PCIe 3.0 slots
  • Two Intel Xeon Scalable CPUs and 4TB memory
  • 480TB storage with sixteen 2.5" drives
  • MIL-STD: 810G, redundant power


RES AI XR6 3U, 8 drive 21 Inch

  • Up to four NVIDIA Tesla V100 or GV100 GPUs with 6 PCIe 3.0 slots
  • Two Intel Xeon Scalable CPUs and 4TB memory
  • 240TB storage with eight 2.5" drives
  • MIL-STD: 810G, redundant power




  • NVIDIA GPGPU accelerators
  • E5-2600 v3/v4 Series Intel® Xeon® processors
  • Up to 1.5TB of DDR4 ECC memory

Enhanced Reliability for Field Operation

Built from the ground up to solve challenges previously reserved for the datacenter, field-deployable RES AI incorporates innovative patented technologies and thermal and mechanical design features to withstand shock, vibration, dust, sand, and temperature extremes. To ensure uptime, availability, and sustained optimal performance in any environment, servers are certified to multiple military (MIL-STD) and commercial (IEC) environmental specifications. Partnered with technology leaders such as Intel, NVIDIA and Mellanox, Mercury maximizes system longevity to support long-term land, naval and airborne operations.

Fully Configurable to Your HPC Application 

Available in multiple sizes and fully configurable, RES AI accelerates an array of high demand workloads by tailoring to unique space, performance, speed and storage requirements. Equipped with numerous PCIe 3.0 slots that accommodate a mix of NVIDIA GPUs, FPGA accelerators, and other expansion cards, servers maximize scalability and minimize the overall computing footprint by consolidating multiple capabilities into a single platform. In order to keep pace with evolving data-driven applications, our supercomputers support the latest SAS3, SATA and NVME drive variations. Compatible with multiple operating systems, applications and software, RES AI scales supercomputing from the Cloud to the Edge. 


Application Highlight: Underwater Electronic Warfare

The processing backbone for submarine electronic warfare, our RES AI 3U server is used to intercept radar and communication signals, trigger threat warnings, identify the targets, and present countermeasures. In order to minimize lag time and simultaneously tackle large amounts of data, each server employs four data center NVIDIA Tesla GPUs, merging the strengths of both CPU and GPU processing capabilities to increase throughput and accelerate performance.  

Safeguard Against Current and Emerging Threats

For security-imperative mission critical applications, servers can be customized with U.S. designed and manufactured motherboards and a variety of advanced security features that safeguard critical IP. When configured with Mercury’s self-encrypting, defense-grade solid state drives (SSD), RES AI delivers data-at-rest protection. A multi-platform compatible, secure hypervisor option that guards against cyberattacks and reverse engineering, is also available.

A Trusted Platform Provider

Mercury servers are designed in the U.S. and manufactured in AS9100D and ISO9001 facilities to maintain stringent quality standards. Committed to protecting information and safeguarding against potential cyberattacks, Mercury complies with DFARS 252.204-7012 and NIST SP 800-171A. With over 30 years of technical expertise, we work closely with customers to design computing solutions that are easy to integrate, affordable and reliable for years to come.