NVIDIA announced its new Tesla 20-series of GPUs for the high performance computing (HPC) market. The Tesla 20-series is based on the company’s new generation CUDA processor architecture, codenamed “Fermi”.
The Tesla 20-series includes the Tesla C2050 and Tesla C2070 GPU Computing Processors, which is for workstation configurations and the Tesla S2050 and S2070 for for cluster and datacenter deployments. The former two are single GPU PCI-Express Gen-2 cards with up to 3GB/6GB of GDDR5 memory offering double precision performance in the range of 520GFlops – 630 GFlops while the latter two are Four Tesla GPUs in a 1U system product with up to 12GB/24GB GDDR5 memory providing 2.1 TFlops – 2.5 TFlops performance.
The Tesla C2050 and C2070 products will retail for $2,499 and $3,999 and the Tesla S2050 and S2070 will retail for $12,995 and $18,995. They will be available in Q2 2010.
The Tesla 20-series GPUs combine parallel computing features that have never been offered on a single device before. These include:
- Support for the next generation IEEE 754-2008 double precision floating point standard
- ECC (error correcting codes) for uncompromised reliability and accuracy
- Multi-level cache hierarchy with L1 and L2 caches
- Support for the C++ programming language
- Up to 1 terabyte of memory, concurrent kernel execution, fast context switching, 10x faster atomic instructions, 64-bit virtual address space, system calls and recursive functions