High Performance Computer Clusters and Technical Computing Workstations
Please note, this list of systems is for your perusal and reference and represent a portion of the technical/HPC workstation and server platforms we have available.
Enterprise Grade NVIDIA GPU Accelerated Workstations and Servers
Dual Socket Intel Xeon E5/V4 Workstation(rackable)
GPU Accelerated Workstation 1x to 4x GPU's
nScale 704 Tesla K40 1x to 4x GPU's
nScale 704 Tesla K80 1x to 4x "
nScale 704 Tesla P100 1x to 4x "
nScale 704 P40 1x to 4x "
nScale 704 P4 1x to 4x "
nScale 704 Intel Phi 3120P 7120 "
nScale 704 AMD FirePro Series "
1~6way PCIe GPU rack server
Dual Socket Intel Xeon E5/V4 HPC server.
nScale 202 Tesla K40 1x to 4x GPU's
nScale 202 Tesla K80 1x to 4x "
nScale 202 Tesla P100 1x to 4x "
nScale 202 P40 1x to 4x "
nScale 202 P4 1x to 4x "
GPU Server 1~8 GPU rack server
nScale 707 Tesla K40 1x-8x GPU's
nScale 707 Tesla K80 1x-8x "
nScale 707 Tesla P100 1x-8x "
nScale 707 P40 1x-8x "
nScale 707 P4 1x-8x "
nScale 707 NVIDIA Titan X or GTX 1080 Retail GPU's
nScale 707 AMD FirePro Cards
nScale 707 FPGA Cards
Server 1~10 GPU rack server
nScale 402 Tesla K40 1x-8x GPU's
nScale 402 Tesla K80 1x-8x "
nScale 402 Tesla P100 1x-8x "
nScale 402 Tesla P40 1x-8x "
nScale 402 Tesla P4 1x-8x "
1~4 way PCIe and NVLink System
nScale 102 1U Dual Xeon E5 V4 Server
-Includes 4x Telsa P100 NVLink* GPU's
IBM S822LC HPC Server
Dual Power8 CPU's and 4x P100 NVLink* 80GB/s CPU to GPU interconnect
nScale 3UX Server 2~16 GPU Expansion Shelf
Connection to host server with Host Interface Connection "HIC" cards
Ships with GTX 1080 cards
nScale 4UX Server 2~16 GPU Expansion Shelf
Connects to host server with Host Interface Connection cards
Ships with your choice of Tesla K80,P100,P40,P4 GPU Accelerators
For NVIDIA's DGX1 with 8x P100 and NVLink, see our dedicated DGX1 page.<
Net Direct authorized (NPN) NVIDIA Preferred Partner since 2008
Net Direct is a leading vendor supplier of High-Performance Computing HPC solutions and clusters in Canada.
Contact us at email@example.com to obtain a custom configured workstation, server or cluster optimized for your specific research or commercial application workload and budget.
*NVLink is a communications protocol developed by Nvidia. NVLink specifies a point-to-point connection between a CPU and a GPU and also between a GPU and another GPU rated at 80 GB/s delivering up to 4x the throughput than PCIe channels.