Nvidia QUADRO PLEX 2200 S4 User manual

Images courtesy of Schlumberger, CyberHeart Inc., WTVF station and WSI corporation, and Realtime Technology AG. QUADRO PLEX 2200 S4 | DATASHEET | OCT08 | FINAL
NVIDIA QUADRO PLEX 2200 S4
ULTRA DENSE REMOTE
GRAPHICS SOLUTIONS QUADRO PLEX DATASHEET
The size and complexity of data is growing at an
exponential rate. In an increasingly competitive
and high-pressure landscape they need to
deliver results better, faster, and more cost
effectively than ever before. These pressures
require the most advanced platform solutions.
The NVIDIA®Quadro®Plex 2200 S4 visual
computing system (VCS), a standard 1U
form factor integrating four ultra-high end
GPUs, delivers the industry’s most advanced
visualization platform for remote and distributed
graphics applications. Professionals ranging
from manufacturing designers and stylists to
earth scientists to digital content creators can
solve their most complex, graphics-intensive
problems using an unconstrained dedicated
visual computing system based on proven,
industry standard architectures.
As a multi-GPU visual computing system,
NVIDIA®Quadro®Plex 2200 S4 dedicated VCS
enables breakthrough performance for all
leading visualization applications. Architected
with four Quadro FX 5800 GPUs, featuring 16
GB of total GPU memory (4 GB/gpu) and 960
CUDA parallel processing cores, the Quadro
Plex 2200 S4 enables graphics-intensive, high-
density visual computing and scales to meet
the most demanding professional applications
requirements.
PRODUCT SPECIFICATIONS
FORM FACTOR
1U Server>
NVIDIA QUADRO GPU
Quadro FX 5800>
# OF NVIDIA QUADRO GPUS
4>
TOTAL FRAME BUFFER
16 GB (4 GB/GPU)>
SHADER MODEL
4.0>
CUDA™PARALLEL PROCESSING CORES
960 (240 /GPU)>
NVIDIA®CUDA™TECHNOLOGY
Yes>
FSAA MAX PER CHANNEL
32x FSAA>
HOST CONNECTION
0.5M Quadro Plex Interconnect Cable>
Standard (option 2.0M cable)
POWER
1200W Max>
ACOUSTICS
Idle 45 dB, max 72 dB>
WEIGHT
~40.0 lbs>
Whether searching for and extracting oil, designing and bringing the
next luxury vehicle to market, or providing a diagnosis of a patient’s
condition, professionals are faced with a mountain of data that needs
to be distilled into meaningful and actionable visualizations.

Where to Buy | www.nvidia.com/quadroplex
© 2008 NVIDIA Corporation. NVIDIA, the NVIDIA logo, NVIDIA Quadro are trademarks or registered trademarks of NVIDIA Corporation.
All rights reserved. All companyand product names are trademarks or registered trademarks of the respective owners with which they
are associated. Features, pricing, availability, and specifications are all subject to change without notice.
SUPPORTED PLATFORMS
Linux>®(64-bit)
Red Had Enterprise Linux 3, 4 and 5•
SUSE 10.1, 10.2 and 10.3•
Microsoft>®Windows®(64-bit)
Windows XP•
Server 2003•
Server 2008•
NVIDIA QUADRO GPU
ARCHITECTURE
128-bit color precision>
(IEEE fp32 bit per component)
3D volumetric texture support>
Fully programmable GPU (OpenGL 3.0 />
OpenGL 2.1 /DirectX 9.0c / DirectX 10)
Shader Model 4.0>
NVIDIA>®CUDA™enabled
C-Programming Environment
DISPLAY RESOLUTION SUPPORT
Hardware Interface Card is designed to>
provide basic console level graphics to one
display
Analog displays up to 2560 x 1600>60 Hz
Single-link DVI-I output - 1280 x 1024>
60 Hz
PRODUCT DETAILS
Quadro Plex Rack Mount Graphics Server>
Standard 19”, 1U rack-mount chassis>
Connects to host via cabling to a low>
power PCI Express x8 or x16 adapter card
Configuration: 2 PCI Express connectors>
driving 2 GPUs each (4 GPUs total)
FEATURES AND BENEFITS
BREAKTHROUGH
VISUAL COMPUTE DENSITY
Unmatched graphics compute per cubic centimeter provides highest visual
compute density enabling breakthrough levels of capability and productivity.
NVIDIA UNIFIED GPU ARCHITECTURE Industry’s first unified architecture designed to dynamically allocate
compute, geometry, shading and pixel processing power to deliver optimized
GPU performance.
HIGH SPEED, PCI EXPRESS
DATA TRANSFER
With low latency and high bandwidth, computing applications benefit from
the highest data transfer rate possible through standard PCI-Express x16
and x8 architecture.
16 GB TOTAL FRAME BUFFER 4 GB/GPU Delivers high throughput for interactive visualization of large models and
high-performance for real time processing of large textures and frames and
enables the highest quality and resolution full-scene antialiasing (FSAA).
STANDARD 1U SERVER FORM FACTOR Industry standard form factor optimized for large scale server deployments.
Four Quadro GPUs in a high density 1U chassis offer the highest performance
for remote graphics applications. Performance optimized and power optimized
products cover the range of IT server room requirements. System monitoring,
thermal control and fault notification in the 1U server product provide the
necessary features for efficient integration of the Quadro Plex 2200 S4 servers
into the data center.
NVIDIA®CUDA™PARALLEL
COMPUTING PROCESSOR
A parallel computing processor architecture with 960 cores (240 / GPU) is
exposed through a C language environment and tool suite in combination with
high performance visualization, CUDA unleashes new capabilities to solve
highly complex challenges such as real-time ray tracing, video encoding, and
interactive volume rendering.
HIGHEST COLOR FIDELITY 10-bit per component color fidelity enables billions rather than millions of
color variations for rich, vivid image quality with the broadest dynamic range.
PCI EXPRESS 2.0 COMPLIANT Maximizes bandwidth between the host processor and the Quadro Plex with up
to 12.8 GB/s transfer rates (up to 6.4 GB/s per PCI Express connection).
TECHNICAL SPECIFICATIONS