Designed from the ground up for ultimate customer satisfaction, the WhisperStation™-Tesla Personal Supercomputer integrates NVIDIA Tesla C1060 or Tesla Fermi C2050/C2070 GPUs. Tesla's massively multi-threaded, parallel-computing architecture, the CUDA™ C and Fortran language environments, and OpenCL support provide the utmost performance for your application.


Link together as a low priced, multi-node, quiet cluster solution.

• Single or Dual Quad/Six-Core Intel Xeon Nehalem or Eight/Twelve-Core AMD Opteron Processors
• Up to 4 Tesla C1060 or Tesla Fermi C2050/C2070 Computing Processors; up to 4 TeraFLOPS of performance!

  Memory:

  up to 144GB of DDR3 1333Mhz ECC Registered DIMMs (Xeon)

  up to 128GB of DDR3 1333MHz ECC Registered DIMMs (Opteron)

• Storage: Up to 5 internal SATA or SAS drives, optionally configured for RAID. Max total storage is 10 TB using 2 TB drives!
• Dual embedded Gigabit Ethernet
• Removable Storage Options: (1) 3.5" floppy drive; (2) DVD/CDRW, DVD-RW, Blu-Ray drives
• I/O features: Front access USB connectors
• Single 1000W power-supply (1350W also available)

• Cuda Best Practices (PDF)
• Volume I: Introduction to CUDA Programming (PDF)
• Volume II: CUDA Case Studies (PDF)

• Excercises (TAR - Linux and MAC)
• Excercises (ZIP - Windows with Visual Studio)
• Excercise Instructions (PDF)

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

KEY BENEFITS of C2050/C2070:

• Up to 448 floating point processor cores
• Up to 1/20th the power consumption and 1/10th the cost of equivalent CPU solutions.
• Tesla GPU's CUDA cores are designed from the ground up for HPC
• Over 1 TFLOP Single and up to 515 GFLOPs Double Precision Floating Point throughput
  • IEEE 754 single & double precision floating point units
• Up to 6GB high speed GDDR5 memory with ECC function
  • Register files, L1/L2 caches, shared memory, and DRAM all are ECC protected.
  • 384-bit memory interface with 144 GB/sec memory bandwidth
• Accessible programming environments
  • Native NVIDIA CUDA C/C++/Fortran environment
  • OpenCL and DirectCompute capable