Overview
PETASCALE COMPUTING WITH TERAFLOP PROCESSORS
ServMax PSC-1, integrated with Tesla C1060 computing processor enables the transition to energy efficient parallel computing power by bring the performance of a small cluster to a workstation. Powered with Intel® Core™ i7 processor and supporting up to 720 processor cores and standard C compilers that simplify application development, ServMax PSC-1 scales to solve the world's most important computing challenges-more quickly and accurately.
Features and Benefits
- Industry's first massively multi-threaded architecture with 240-cores.
- Many-core architecture delivers optimum scaling across HPC applications.
- Optimized for scientific computing, delivering up to 15x cost savings and 15x lower power than traditional 1U rack-optimized servers.
- Scale to thousands of processor cores to solve large-scale problems by splitting the problem across multiple GPUs.
- High-efficiency computing platform for energy-conscious organizations.
- NVIDIA CUDAT Technology unlocks the power of Tesla many-core computing products.
- Seamlessly able to fit into existing HPC environments.
- Ideal for life sciences, geosciences, engineering & sciences, molecular biology, medical diagnostics, electronic design automation (EDA), government and defense, visualization, financial modeling, and oil & gas applications.
Specification
- Up to three Tesla processors (240 computing cores per processor, 720 cores total)
- Delivers up to three teraflops in a tower chassis with a typical energy footprint of 1,000 watts
- IEEE 754 single & double floating point precision
- Supports single Intel® 45nm Multi-Core CPU / Intel® Core™ i7 processor
- Up to 12 GB dedicated memory (organized as 4.0 GB per GPU)
- 3 x 512-bit GDDR3 memory interface (organized as a 512-bit interface per GPU)
- Up to 306 GB/sec memory bandwidth (102 GB/s per GPU to local memory)
- Ultra quiet, eloquent mid tower chassis
- System dimension: 23.6" x 9.6" x 24.6" (D x W x H)
|
1U Rackmount
