- Delivers up to 515 Gigaflops of double-precision peak performance in each GPU, enabling a single workstation to deliver a Teraflop or more of performance
- Single precision peak performance is over a Teraflop per GPU
- Meets a critical requirement for computing accuracy and reliability for workstations
- Maximizes performance and reduces data transfers by keeping larger data sets in local memory that is attached directly to the GPU
- Choose C, C++, OpenCL, DirectCompute, or Fortran to express application parallelism and take advantage of the ""Fermi"" GPU's innovative architecture
The NVIDIA Tesla 20-series makes supercomputing available to everyone. Based on the massively parallel CUDA architecture, it features many "must have" HPC capabilities including ECC memory for uncompromised accuracy and reliability, an L1/L2 cache for improved bandwidth and data sharing, larger dedicated compute memory, support for C++, and 7X double precision performance of the previous generation. The Tesla 20-series delivers performance equivalent to a CPU-only system at 1/10th the cost and 1/20th the power consumption.