Microway and NVIDIA are offering a 50% discount on the Tesla C1060 Compute Processor until July 24. Developers may purchase up to four C1060s at the special price of $650 + $25 shipping and handling (U.S.). To receive this discount, please fill out the form below and send your NV discount code to sales@microway.com:
http://www.nvidia.com/object/tesla_promo_apr09_Microway.html

Microway HPC Clusters and Whisperstations seamlessly integrate the NVIDIA^® Tesla™
high performance compute processors. Solve your most important HPC challenges with a massively multi-threaded, parallel-computing architecture and the CUDA™ C and Fortran language environment and tool suites.

Microway Tesla Products Data Sheet (PDF)

Jacket is a GPU engine for MATLAB^®, which enables standard MATLAB code to run on the GPU. Sign up for free 15 day trial.

Tesla GPU computing solutions also fit seamlessly into your existing workstation or HPC infrastructure to enable you to solve problems orders-of-magnitude faster.

KEY BENEFITS:

• Massively multi-threaded architecture with 240 floating point processor cores
• IEEE 754 Floating Point [single and double-precision] - meets the precision requirements of your most demanding applications with 64-bit ALUs
• 933 GFLOPS Single and 78 GFLOPs Double Precision Floating Point throughput
• 4 GB High-Speed Memory per processor (102 GB/s transfer)
• Fits seamlessly into existing HPC environments

Tesla C1060 GPU Card

The C1060 GPU Computing Processor transforms a Microway WhisperStation into a personal supercomputer system with one teraflop of peak floating point performance.

Tesla S1070 GPU Server

The S1070 GPU Computing Server seamlessly fits into HPC server clusters and scales to solve the most complex computing problems. Connects to host(s) via PCI-E x16 2.0 or x8.

Massively Parallel Performance
A massively multi-threaded processor mines large datasets to extract information and reach answers fast. Key features available on Tesla GPUs include the Thread Execution Manager to coordinate the concurrent execution of thousands of parallel computing threads and the Parallel Data Cache to enable those threads to share data easily.



C for the GPU - NVIDIA® CUDA™
The world's only C-language development environment for the GPU, the NVIDIA® CUDA™ software development kit includes a standard C compiler, hardware debugger tools, and a performance profiler for simplified application development.



Fortran for the GPU - The Portland Group (PGI) CUDA™ Compiler
In June 2009 PGI and NVIDIA announced cooperation to develop CUDA Fortran. CUDA Fortran includes a Fortran 2003 compiler and tool chain for programming NVIDIA GPUs. PGI will support CUDA Fortran with a release of its Fortran compiler later this year.




Power to Solve Complex Problems
Scientists and researchers benefit from the power of the massively parallel, computing architecture in Tesla solutions. This availability of supercomputing will unlock the answers to previously unsolvable problems in systems ranging from a workstation to server clusters.



Compatible Solutions
As an industry-standard solution, Tesla easily fits into existing HPC environments. Available products include a GPU for users to upgrade their existing workstation, a deskside system to add additional performance alongside a workstation, and a 1U GPU server for deployment within an enterprise data center.



Developer Community
NVIDIA is the catalyst for the largest GPU computing developer community. NVIDIA's interactive, on-line GPU developer community provides access to forums, educational materials, and additional resources and tools.

New: 1U Compute Node with Two Tesla C1060 GPUs

Microway's Tesla Personal Supercomputer with
2 Tesla C1060 Compute Processors

Microway's Tesla GPU + InfiniBand Cluster with
36 Quad-Core CPUs, 9 S1070s with 36 Tesla 1TFLOP GPUs
and 40 Gbit/sec QDR ConnectX InfiniBand