Quadro in the Data Center

From data science to virtual environments, enterprises are tackling larger, more complex workloads than ever before. Meet the most demanding visual computing challenges by bringing the power of NVIDIA Quadro RTX GPUs and NVIDIA virtual GPU software to the data center. With advanced technology for AI, real-time ray tracing, and graphics, IT teams can deploy servers capable of a wide range of workloads at a fraction of the cost, space, and power requirements of CPU-based solutions.

The ray-tracing capabilities of NVIDIA RTX help artists push the boundaries of content creation. Arnold GPU is now available and we’re excited to see how our users will take advantage of the unprecedented rendering speeds and amazing efficiency.

— Chris Vienneau, Senior Director of Media and Entertainment Products, Autodesk

Cinesite was proud to be an early partner with NVIDIA on the RTX platform, but we never expected results this dramatic. We can iterate faster, more frequently, and with higher quality. This will completely change how our artists work.

— Michele Sciolette, CTO, Cinesite

One of the biggest challenges in virtual production is achieving physically accurate lighting in real time. RTX enables real-time ray tracing within our Genesis platform, making photorealism possible for the first time ever live on set.

— Francesco Giordana, Realtime Software Architect, MPC Film

Accelerate Data Center Workloads

Tackle the most graphics- and compute-intensive workloads with Quadro RTX-powered servers.

Virtual Workstations

Virtual Workstations

  • Access all of the benefits of the Quadro® platform in a virtual environment from the data center with NVIDIA Quadro Virtual Data Center Workstation (Quadro vDWS) software.
  • Tackle massive datasets, large 3D models, and complex designs with the scaled memory and performance of Quadro RTX 6000 or RTX 8000 GPUs with Quadro vDWS software.
Rendering
Image courtesy of Neoscape and Design by Safdie Architects.

Rendering

  • NVIDIA® NVLink delivers up to 96 gigabytes (GB) of GPU memory for IT-ready, purpose-built Quadro RTX GPU clusters that massively accelerate batch and real-time rendering in the data center.
  • Through the combination of RT Cores and Tensor Cores, the RTX platform brings real-time ray tracing, denoising, and AI acceleration, letting artists create photorealistic renders at a fraction of the cost, space, and power requirements of CPU-based render farms.
AI and Data Science

AI and Data Science

Mixed Workloads

Mixed Workloads

  • Host virtual machines running diverse graphics, deep learning, and simulation workloads with a common data center infrastructure running on NVIDIA Quadro RTX GPUs with NVIDIA virtual GPU software
  • Run mixed workloads in servers to increase productivity, maximize GPU utilization and efficiency, and reduce data center costs.
Augmented and Virtual Reality Over 5G

Augmented and Virtual Reality Over 5G

  • Accelerate edge computing for graphics-intensive workloads with NVIDIA Quadro vDWS.
  • Deploy servers across 5G networks, leveraging the high bandwidth and low latency to create immersive augmented and virtual reality experiences.
Simulation and Scientific Visualization

Simulation and Scientific Visualization

  • Run large-scale simulations quickly and accelerate time to scientific discovery with GPU-powered parallel processing.
  • Combine Quadro GPUs and NVIDIA virtual GPU software to scale workloads according to size and complexity.

Built for Professionals

Bringing Quadro to the data center expands beyond groundbreaking hardware—it gives IT managers a complete visualization platform to simplify even the most complex deployments.

  • Quadro long-life drivers are engineered by NVIDIA and extensively tested and certified by major workstation OEMs to minimize interruptions and maximize productivity.
  • Certifications for over 100 professional applications guarantee hardware stability and performance.
  • Quadro vDWS software virtualizes any graphics application from the data center for increased security and easier management.
  • High-end Quadro graphics cards feature both active and passive thermal cooling to fit a variety of servers.
  • The NVIDIA Enterprise Management Toolkit (NVWMI) increases operational efficiency by easily and remotely deploying, configuring, updating, and monitoring Quadro solutions.
  • Enterprise-grade support from NVIDIA Quadro experts and long-life warranties deliver peace of mind.
NVIDIA RTX SERVER

NVIDIA RTX SERVER

Quadro-powered servers can transform your data center infrastructure. Take visual computing to the next level with NVIDIA RTX Server, a highly flexible reference design that features OEM server systems that have been tested and validated across a range of workloads. NVIDIA RTX Server-validated systems combine NVIDIA Quadro RTX GPUs with NVIDIA virtual GPU software and leading third-party applications for maximum performance and cost savings.

Powered by the World's Most Advanced GPUs

  • Passive Thermal Solution
  • Active Thermal Solution
RTX 8000

Resources

Specifications

NVIDIA CUDA Cores 4,608 CUDA Cores
Tensor Cores 576 Cores
RT Cores 72 Cores
Performance with NVIDIA GPU Boost NVIDIA CUDA Cores
Interconnect Bandwidth (Bidirectional) 100 GB/s
Memory 48 GB GDDR6
Power Consumption 250 Watts
Thermal Solution Passive
Displays Connectors No physical display connectors.*

Display output requires Quadro vDWS license.
To learn more visit www.nvidia.com/en-gb/design-visualization/quadro-vdws/

Resources

RTX 8000

Resources

Specifications

NVIDIA CUDA Cores 4,608 CUDA Cores
Tensor Cores 576 Cores
RT Cores 72 Cores
Performance with NVIDIA GPU Boost NVIDIA CUDA Cores
Interconnect Bandwidth (Bidirectional) 100 GB/s
Memory 48 GB GDDR6
Power Consumption 250 Watts
Thermal Solution Active
Displays Connectors 4xDP 1.4, 1x USB-C

Resources

  • Passive Thermal Solution
  • Active Thermal Solution
RTX 6000

Resources

Specifications

NVIDIA CUDA Cores 4,608 CUDA Cores
Tensor Cores 576 Cores
RT Cores 72 Cores
Performance with NVIDIA GPU Boost NVIDIA CUDA Cores
Interconnect Bandwidth (Bidirectional) 100 GB/s
Memory 24 GB GDDR6
Power Consumption 250 Watts
Thermal Solution Passive
Displays Connectors No physical display connectors.*

Display output requires Quadro vDWS license.
To learn more visit www.nvidia.com/en-gb/design-visualization/quadro-vdws/

Resources

RTX 6000

Resources

Specifications

NVIDIA CUDA Cores 4,608 CUDA Cores
Tensor Cores 576 Cores
RT Cores 72 Cores
Performance with NVIDIA GPU Boost NVIDIA CUDA Cores
Interconnect Bandwidth (Bidirectional) 100 GB/s
Memory 24 GB GDDR6
Power Consumption 250 Watts
Thermal Solution Active
Displays Connectors 4xDP 1.4, 1x USB-C

Resources

Quadro OEM Servers

Advanced HPC
Amax
ASRock Rack
Asus
Boxx
Colfax
Dell EMC
Equus Compute Solutions
Exxact
Fujitsu
Gigabyte
GPL Technologies
Hewlett Packard Enterprise
Hitachi
Images & Technologie
Inspur
Lenovo
Penguin Computing
Quanta Computer
Rave Computer
Supermicro

Find an NVIDIA Data Center GPU Partner

Find a vGPU Certified Server Partner