Supermicro New Zealand
Tesla C2075 and NVIDIA Maximus Technology
  29 February 2012

In Q4 of 2011 NVIDIA released the Tesla C2075 co-processor, designed to offload parallel computations from the CPU to drastically accelerate application performance. Engineers, designers and content creation professionals are already enjoying the benefits of NVIDIA GP-GPU capabilities with many industry software solutions available today. The list of software that benefit from CUDA is growing and will continue to grow - see a sample list in the table below. It is a safe prediction that we will eventually have more software taking advantage of the GPU to speed up general computing performance for the average business user and consumer as well.

Field
ISV/Application
 Molecular Dynamics Abalone, ACEMD, AMBER, DL-POLY
 Quantum Chemistry
 		TeraChem, VASP
 Bioinformatics GPU-Blast, PIPER Protein Docking, SeqNFind, UGene, CUDASW++
 Computational Fluid Dynamics Altair AcuSolve, Autodesk Moldflow, FluiDyna Culises for OpenFOAM, Prometech Particleworks, Turbostream Ltd.
 Computational Structural Mechanics ANSYS Mechanical, Abaqus, Impetus, MSC Nastran
 Computational Finance Hanweck Associates, Murex, MathWorks MATLAB, Numerical Algorithms Group
 EDA and CEM Agilent Technologies EMPro & ADS, ANSYS Nexxim, CST Microwave Studio (MWS), Gauda OPC, OPV
 Weather and Climate Forecasting
NIM (NOAA), WRF (NCAR)
 Oil and Gas GeoStar RTM, GeoMage Multifocusing, PanoramaTech: Marvel, Paradigm Echos, SKUA, VoxelGeo, Seismic City RTM, Tsunami A2011
 Medical Imaging Acceleware AxRecon, Digisens
 Data Mining AccelerEyes Jacket for MATLAB, MathWorks MATLAB, Wolfram Mathematica, Jedox Palo, ParStream, GPU-Quicksort, GPU-LIBSVM
 Numerical Analysis AccelerEyes Jacket for MATLAB, MathWorks MATLAB, Wolfram Mathematica

Of more importance is the release of NVIDIA's Maximus technology. NVIDIA Maximus-powered workstations by Compucon combines the professional 3D capability of the Quadro with the high performance computing power of the Tesla. Essentially this allows each of the cards to focus on what they do best; real-time interactive graphics or CAE rendering/computations.

Visualising and simulating at the same time? For engineers, artists, designers or scientists this means you can now interact with high-performance visuals while performing simulations or renderings on the same system – at the same time. This changes the slow traditional workflow of design then simulation (by the CPU) into much faster and shorter iterations of design and simulation cycles (by the GPU).

On the other hand, Maximus also allows this Quadro + Tesla arrangement to combine and produce excellent performance and more bang for your buck than ever before. In fact, the secret is that with Maximus technology, a Quadro 2000 card combined with a Tesla C2075 provides greater performance per dollar than a single Quadro 6000 card as benchmarked by NVIDIA under Adobe Premier Pro.


Image
Source: NVIDIA
Please contact us to see if you can benefit from this latest technology and/or class of products.

Editor

###

Supermicro New Zealand
Managed by Compucon New Zealand