NVIDIA Tesla K80.

NVIDIA Tesla K80.

Description:

Speed ​​up the processing of the most demanding computing tasks with single and double precision in the field of scientific research, observations of seismic activity, as well as applied data analysis by switching to the use of the NVIDIA Tesla K80 computing accelerator with two graphic processors. This accelerator provides increased productivity up to 2.2 times compared to the Tesla K20X series, two and a half times higher performance compared to the Tesla K10 models and a tenfold advantage over modern central processors when performing applied computing tasks.
The Tesla K80 accelerator provides a more than twofold increase in application speed compared to the previous generation of accelerators, as well as performance up to 10 times higher than central processors. Featuring unique features such as 24 GB GDDR5 memory with 480 GB / s bandwidth and advanced Boost GPU technology, the Tesla K80 Series delivers processing power that will allow you to process petabytes of data and computer simulation tasks faster than ever before.

Features and Specifications:

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

PNY Part Number - TCSK80M-PB

Available extender brackets:    

     Long Extender with Offset

     Short Extender with Offset.

Package Including:

1 x NVIDIA Tesla K80.

You can see official accounts our company:

Made-in-China: http://www.made-in-china.com/showroom/beiker888

Alibaba: https://cn1500459308.onesite.alibaba.com

For more details and wholesale catalog, please contact our managers.

Wholesaler - special rates and conditions.

sales@china-sells.com

Waiting for your orders!