Memory Size 24GB GDDR5 (12 GB per GPU)

Memory Interface 384-bit

Memory Bandwidth (ECC off))480GB/s (per board)240 GB/s (per GPU)

CUDA Cores 4992

System Interface PCI Express 3.0 x16

Processor core clock Base clock: 560 MHz                               Boost clock: 562 - 875 MHz

Memory clock 2.5 GHz

Cache L1 and L2

Max. Consumption 300 W 

Number of GPUs 2 × GK210

Energy Star Enabling Yes

Conformité Européenne (CE) Yes

Thermic Solution Passive

Format 111.150 mm (H) × 267 mm (L) - Dual Slot, Full-Height

Connectors 1 × 8-pin CPU power connectors                 (2 x 8-pin PCIe to single 8-pin CPU adapter included)

SMX Yes

Dynamic Parallelism Yes

Hyper-Q Yes

GPU Computing Applications Reservoir simulation                                         CAE (structural analysis)                                         Molecular dynamics, Numerical analytics                                         Computational visualization (ray tracing)

Peak double precision floating point performance 2.91 Tflops (GPU Boost Clocks)                                                                       1.87 Tflops (Base Clocks)

Peak single precision floating point performance 8.74 Tflops (GPU Boost Clocks)                                                                      5.6 Tflops (Base Clocks)

System Servers 

Compatible with Windows  64-bit                        Linux 64-bit  P

NY Part Number TCSK80M-PB

Available extender brackets