GPU Optimised Workstations

For the ultimate in number crunching, our GPU workstations can be configured with up to 6x Intel Phi's or Teslas

  • Configurable with Intel Xeon Phi or nVidia Tesla GPU coprocessors
  • Enterprise-Grade Server Components
  • GPU / Supercomputing Technical Sales Team
  • 3 Years Warranty Included with Extended Warranty Options Available
  • Dual E5-2600 v3 Processors


Matching Workstations 2

SystemFeaturesDrive BaysExpansion SlotsWorkstation SizeProcessorMax RAM Capacity
 
TeslaStation Pro Intel X99, supports 4-Way Geforce SLI and Quad CrossFireX and Telsa, 8 x DIMM, Max. 128GB Quad Channel Memory, Dual Intel Gigabit LAN, 8-Channel High Definition Audio CODEC featuring Crystal Sound 2.
Configure
from: £1,408.09

Ultra-Quiet Design

Multi-Display Options

Tesla, Phi, GTX GPU

4 Hard Drives


7 x PCIe 3.0/2.0 x16

Full Tower Large Workstation

Intel Core i7 Extreme Processor

128GBMax RAM

TeslaStation Pro-XL Dual Xeon GPU Computing Pedestal Supercomputer, 4x Tesla, Xeon Phi or GTX-Titan GPU Cards
Configure
from: £2,599.69

Ultra-Quiet Design

Tesla, Phi, GTX GPU

8 Hot-Swap Hard Drives


4 x PCIe 3.0/2.0 x16, 2 x PCIe 3.0/2.0 x8, 1 x PCIe 3.0/2.0 x8(x4 Speed)

Full Tower Large Workstation

Xeon E5 v4 Processor

512GBMax RAM

Call a Broadberry Storage & Server Specialist Now: 020 8997 6000

Have a Broadberry Expert Contact You:

The Broadberry CyberStation GPU® range of supercomputing workstations delivery 20x higher application performance than the latest multi-core CPU systems.

Built to provide unprecedented real-world scalability and breakthrough power-efficiency across a wide array of HPC applications using the latest enterprise-grade workstation technology with up to 6x Intel Xeon Phi or nVidia Tesla coprocessors.

CyberStation GPU Workstations:

Ultra fast GPU supercomputers for when processing performance is key, these enterprise-grade workstations are built to solve the world's most complex computational problems

  • Ultra-Performance Enterprise Supercomputing
  • 12GB 2.5" / 3.2" SAS / nVME Drives
  • nVidia Tesla / Intel Xeon Phi compatible
  • Ultra-High Teraflop Performance

High-Performance, Enterprise-Class Supercomputing GPU-Accelerated for Finance, Oil & Gas, Science, and Engineering

Supporting the latest nVidia Tesla, GTX and Intel Xeon Phi coprocessors, Broadberry GPU workstations are configurable with the latest Graphics Processing Cards for the most complex calculations.

Features:

  • Maximum Bandwidth: PCI-E 3.0 x16 for best I/O performance
  • Manageability: Server management & GPU status monitoring through IPMI 2.0
  • Reliability: Redundant power supplies & intelligent cooling control
  • Efficiency: Breakthrough system architecture to optimise TCO, with Platinum Level power supplies, advanced cooling, and high-end motherboard components

Features Tesla K80 Tesla K40 Titan X
GPU 2x Kepler GK210 1 Kepler GK110B 1x Maxwell
Peak double precision floating point performance 2.91 Tflops (GPU Boost Clocks) 1.87 Tflops (Base Clocks) 1.66 Tflops (GPU Boost Clocks) 1.43 Tflops (Base Clocks) 0.192 Tflops
Peak single precision floating point performance 8.74 Tflops (GPU Boost Clocks) 5.6 Tflops (Base Clocks) 5 Tflops (GPU Boost Clocks) 4.29 Tflops (Base Clocks) 6.2 Tflops
Memory bandwidth (ECC off) 480 GB/sec (240 GB/sec per GPU) 288 GB/sec 288.4 GB/sec
Memory size (GDDR5) 24 GB (12GB per GPU) 12 GB 12 GB
CUDA cores 4992 ( 2496 per GPU) 2880 3072

Delivered application performance is more than raw flops, computational professionals depend on mission critical applications to accelerate discoveries and insights at scale.

It starts with the world's fastest accelerators and includes a reliable infrastructure, the ability to monitor and manage that infrastructure, and the ability to quickly move data where it is needed. The NVIDIA Tesla Accelerated Computing Platform provides all of these to deliver unprecedented performance in scientific, analytics, engineering, consumer, and enterprise applications.

The Worlds Fastest GPU Accelerators
Fast Communication
Reliable Infrastructure Management

Although designed for use as a graphics card, the GeForce GTX Titan X card is reguarly used by many price-savvy customers as a GPU accelerator.

Boasting around the same amount of CUDA cores as Tesla cards, Titan X cards have more CUDA cores than the K40, the same amount of RAM, a higher clock speed and 2 DMA engines for bi-directional data copying - all for around 1/4 of the price of a Tesla K40.

GeForce Titan X Cards:
  • 3072 Cores
  • 12GB Ram
  • 336GB/s, 384-bit bus
  • Boost Clock: 1075MHz
  • Core Clock 1000MHz
  • 7GHz DDR5
  • 6.2 TFLOPS SP
What won't a Titan-X card do as well as a Tesla?
  • Poor double-precision performance
  • Produced by 3rd Party Companies - not nVidia
  • No ECC Memory Error Protection

High-Performance, Enterprise-Class Supercomputing GPU-Accelerated for Finance, Oil & Gas, Science, and Engineering

Intel Xeon Phi coprocessors are PCI-Express expansion cards that work along-side the Intel Xeon processor to enable dramatic performance gains for highly parallel code. Broadberry high-performance GPU servers can accomodate up to 8x Intel Xeon Phi cards - delivering up to 1.2 teraFLOPS per coprocessor.

Manufactured using Intel's Industry-leading 22nm technology with 3-D Tri-Gate transistors, each coprocessor features more cores, more threads and wider vector execution units than an Intel Xeon processor, compensating for the lower speed of each core to deliver higher aggregate performance for highly parallel workloads.to deliver higher aggregate performance for highly parallel workloads.


Configure your Broadberry High-Performance GPU workstation with up to 6 of the Intel Xeon Phi coprocessors below.

Product Name Processor Number Cache # of Cores Processor Base Frequency TDP Max Turbo Frequency Max Memory Size Max # Memory Channels Max Memory Bandwidth
Intel© Xeon Phi Coprocessor 3120A (6GB, 1.100 GHz, 57 core) 3120A 28.5 MB L2 Cache 57 1.1 GHz 300 W 6 GB 12 240 GB/s
Intel© Xeon Phi Coprocessor 3120P (6GB, 1.100 GHz, 57 core) 3120P 28.5 MB L2 Cache 57 1.1 GHz 300 W 6 GB 12 240 GB/s
Intel© Xeon Phi Coprocessor 5120D (8GB, 1.053 GHz, 60 core) 5120D 30 MB L2 Cache 60 1.053 GHz 245 W 8 GB 16 352 GB/s
Intel© Xeon Phi Coprocessor 5110P (8GB, 1.053 GHz, 60 core) 5110P 30 MB L2 Cache 60 1.053 GHz 225 W 8 GB 16 320 GB/s
Intel© Xeon Phi Coprocessor 7120A (16GB, 1.238 GHz, 61 core) 7120A 30.5 MB L2 Cache 61 1.238 GHz 300 W 1.333 GHz 16 GB 16 352 GB/s
Intel© Xeon Phi Coprocessor 7120D (16GB, 1.238 GHz, 61 core) 7120D 30.5 MB L2 Cache 61 1.238 GHz 270 W 1.333 GHz 16 GB 16 352 GB/s
Intel© Xeon Phi Coprocessor 7120P (16GB, 1.238 GHz, 61 core) 7120P 30.5 MB L2 Cache 61 1.238 GHz 300 W 1.333 GHz 16 GB 16 352 GB/s
Intel© Xeon Phi Coprocessor 7120X (16GB, 1.238 GHz, 61 core) 7120X 30.5 MB L2 Cache 61 1.238 GHz 300 W 1.333 GHz 16 GB 16 352 GB/s

While a majority of applications (80 to 90 percent) will continue to achieve maximum performance on Intel Xeon processors, certain highly parallel applications will benefit dramatically by configuring a Broadberry server with Intel Xeon Phi coprocessors.

To take full advantage of Intel Xeon Phi coprocessors, an application must scale well to over 100 software threads and either make extensive use of vectors or efficiently use more local memory bandwidth than is available on an Intel Xeon processor.


Performance

Delivers over a teraFLOPS (floating-point operations per second) of double-precision peak performance. Additionally, as compared with Intel® Xeon® processor E5 family-based workstations, the Intel® Xeon Phi coprocessor delivers up to 2.3 times higher peak FLOPS, up to 3 times more performance per watt, and up to 7 times better performance on certain financial services applications.

Common Programming

Applications can support both Intel® Xeon® processors and Intel® Xeon Phi coprocessors which use common languages, models, and familiar and standard development tools, so there's no need to learn new languages or tools.

Flexible Compute Modes

Leverage the compute flexibility of Intel® Xeon Phi coprocessors to either use in Intel® Xeon® processors to accelerate processing of highly parallel code, or create independent, high-performance computing compute nodes with its own IP address and can run applications independently unlike a basic accelerator.


Our Rigorous Testing

Before leaving our UK workshop, all Broadberry server and storage solutions undergo a rigorous 48 hour testing procedure. This, along with the high-quality industry leading components we use ensure all of our server and storage solutions meet the strictest quality guidelines demanded from us.


Un-Equaled Flexibility

Our main objective is to offer great value, high-quality server and storage solutions, we understand that every company has different requirements and as such are able to offer un-equaled flexibility in designing custom server and storage solutions to meet our clients' needs.

Trusted by the World's Biggest Brands

We have established ourselves as one of the biggest storage providers in the UK, and since 1989 supplied our server and storage solutions to the world's biggest brands. Our customers include: