The Trusted Leader in High Performance Computing


SGI® with NVIDIA® GPU Accelerators

Complete, Factory-integrated GPU Accelerator Solutions


Learn More

SGI offers unique GPU Computing solutions including the latest NVIDIA® Tesla® GPU computing solutions including the Tesla M40, K80 and K40 GPU Accelerators. GPU solutions are certified and available in SGI's complete line of servers: SGI® Rackable®, SGI® UV™, and SGI® ICE™. SGI also provides CUDA training and professional services for turn-key solutions for customers.

Workgroup to Enterprise

  • SGI Rackable Servers
    Leveraging the winning combination of the latest Intel® Xeon® Processor architecture and the NVIDIA® Tesla® GPU Accelerators, these servers deliver top value and performance. SGI Rackable servers are tailored to meet your exact specifications for high-density, high GPU to CPU ratios, high I/O or high memory.

  • SGI UV 30
    A quad-socket Intel® Xeon® server, featuring rich memory and I/O in a compact, 2U footprint, the SGI UV 30 is a great solution for many scientific and engineering workloads. It can be used as a head node or fat node for a scale-out cluster, or standalone. When paired with NVIDIA Tesla GPU Accelerators, the resulting compute and memory density is available for large models and complex optimization.


  • SGI UV
    The UV platform brings GPUs to a new class of solutions in chemistry, homeland defense, fluid dynamics and biosciences. New SGI UV 3000 systems are designed for compute-intensive, fast algorithm workloads such as CAE, genome assembly, and scientific simulations. Newly enhanced SGI UV 300 and SGI UV 30EX servers are designed for data-intensive, I/O heavy workloads such as data analytics, visualization, and real-time streaming.

    SGI ICE XA, a 6th generation supercomputer addressing the increasing size and volume of HPC workloads, provides optimum flexibility in processor and accelerator mix, blade and storage configuration, and network topology to meet specific workload requirements while allowing to grow existing ICE clusters by adding next generation nodes without user interruption.

SGI GPU Compute Solutions at a Glance

Solution Vertical "U" Accelerator
SGI® UV™ 3000 10U blade enclosure 1x NVIDIA® Tesla® per MG blade/16x NVIDIA® Tesla® per SSI
SGI® UV™ 300 5U 4x NVIDIA® Tesla® per 5U chassis/16x NVIDIA® Tesla® per SSI
SGI® UV™ 30EX 5U 4x NVIDIA® Tesla®
SGI® UV™ 30 2U 2x NVIDIA® Tesla®
Rackable C2108-GP5 2U 8x NVIDIA® Tesla®
Rackable C2112-GP2-G 2U 2x NVIDIA® Tesla®
Rackable C1104-GP1 1U 3x NVIDIA® Tesla®
SGI® ICE™ XA 10.5U blade enclosure 4x NVIDIA® Tesla® per IP-139CS blade service node
SGI® UV™ 2000 10U blade enclosure 1x NVIDIA® Tesla® per MG blade/16x NVIDIA® Tesla® per SSI
Rackable C1104G-RP5 1U 3x NVIDIA® Tesla®
Rackable C2110G-RP5-P 2U 3-4x NVIDIA® Tesla®
Rackable C2108-RP2 2U 2x NVIDIA® Tesla®