PNY NVIDIA Tesla K40 GPU Computing Accelerator - 12GB
Solve your most demanding High-Performance Computing (HPC) challenges with NVIDIA Tesla family of GPUs. They're built on the NVIDIA Kepler™ compute architecture and powered by NVIDIA CUDA®, the world's most pervasive parallel computing model. This makes them ideal for delivering record acceleration and more efficient compute performance for big data applications in fields, including seismic processing; computational biology and chemistry; weather and climate modeling; image, video and signal processing; computational finance, computational physics; CAE and CFD; and data analytics.
Equipped with 12 GB of memory, the Tesla K40 GPU accelerator is ideal for the most demanding HPC and big data problem sets. It outperforms CPUs by up to 10x and includes a Tesla GPUBoost feature that enables power headroom to be converted into user controlled performance boost.
The innovative design of the Kepler compute architecture includes:
SMX (streaming multiprocessor)
Delivers up to 3x more performance per watt than the SM in last-generation NVIDIA Fermi GPUs.
Enables GPU threads to automatically spawn new threads, by adapting to the data without going back to the CPU, this greatly simplifies parallel programming.
Allows multiple CPU cores to simultaneously use the CUDA cores on a single Kepler GPU. This dramatically increases GPU utilization and slashes CPU idle times. Features ECC MEMORY ERROR PROTECTION
Meets a critical requirement for computing accuracy and reliability in datacenters and supercomputing centers. Both external and internal memories are ECC protected in Tesla K40.
SYSTEM MONITORING FEATURES
Integrates the GPU subsystem with the host system’s monitoring and management capabilities such as IPMI or OEM-proprietary tools. IT staff can thus manage the GPU processors in the computing system using widely used cluster/grid management solutions.
L1 AND L2 CACHES
Accelerates algorithms such as physics solvers, ray-tracing, and sparse matrix multiplication where data addresses are not known beforehand
ASYNCHRONOUS TRANSFER WITH DUAL DMA ENGINES
Turbocharges system performance by transferring data over the PCIe bus while the computing cores are crunching other data.
FLEXIBLE PROGRAMMING ENVIRONMENT WITH BROAD SUPPORT OF PROGRAMMING LANGUAGES AND APIS
Choose OpenACC, CUDA toolkits for C, C++, or Fortran to express application parallelism and take advantage of the innovative Kepler architecture.
End-user can convert power headroom to higher clocks and achieve even greater acceleration for various HPC workloads on Tesla K40.
|Graphics Chipset||Tesla K40|
|NVIDIA CUDA Cores||2880|
|NVIDIA Tensor Cores|
|Memory Size||12 GB|
|Memory Bit Rate|
|Memory Bandwidth||288 GB/s|
|Single Precision Processing||Yes|
|Single Precision Performance||4.29 teraFLOPS|
|Double Precision Processing||Yes|
|Double Precision Performance||1.43 teraFLOPS|
|Deep Learning (Tensor) Performance|
|Integer Operations (INT8)|
|H.264 1080p30 Streams||N/A|
|Cooling||Single Fan (1)|
|Interface||PCIe 3.0 (x16)|
|Maximum Digital Resolution|
|Maximum VGA Resolution|
|Multi GPU Support|
|Multi Monitor Support|
|Supported Compute APIs|
|Supported Graphics APIs|
|Microsoft DirectX Support|
|Low Profile Compatible||No|
|Low Profile Support|
|Graphics Card Power Connectors|
|Minimum Recommended PSU|
|Maximum GPU Temperature|
Please note your statutory rights are not affected.
For further information regarding Scan's warranty procedure please see our terms and conditions
- 36 months
- DOA Period:
- 28 days
- RTB Period:
- 12 months
Date Issued: 2nd Dec 2010
NVIDIA® CUDA parallel computing architecture is enabled on GeForce, Quadro, and Tesla products. Whereas GeForce and Quadro are designed for consumer graphics and professional visualization respectively, the NVIDIA® Tesla™ product family is designed ground-up for parallel computing and offers exclusive computing features.
Date Issued: 8th Oct 2010
Modern desktop computers and notebooks comprise of a CPU, motherboard, graphics, storage, and, usually an optical drive. Computers have a number of ports and sockets that enable the user to plug-in various peripherals such as a printer, USB mouse, or, perhaps most importantly of all, an Internet connection.
Date Issued: 22nd Oct 2008
CUDA technology is the world’s only C language environment that enables programmers and developers to write software to solve complex computational problems in a fraction of the time by tapping into the many-core parallel processing power of GPUs.
Date Issued: 20th Oct 2008
This TekSpek explains why you’d want to overclock your graphics board, the risks in doing so and how you can go about doing it.
Date Issued: 20th Oct 2008
This TekSpek will assume you know the affects of applying a level of anti-aliasing (AA) on your 3D accelerator, be it via the driver control panel or via a control in your game. We assume you know the effect it has on image quality, so you can think about a before and after scenario. So this TekSpek isn’t about explaining what it does as such, although it will, it’s about explaining the how and why.
Date Issued: 20th Oct 2008
Explaining how a modern GPU works in completeness would take a book. Or two. Per class of chip. Per vendor. They’re extraordinarily complex pieces of engineering and production, and the end result contains more transistors than multiple modern x86 processors.
Date Issued: 25th Jun 2008
This TekSpek explains DirectX 10, which graphics cards support it, how they work and what the consumer’s choices are.
Date Issued: 14th Jun 2008
We’ve all been victims of static electricity at some point. Perhaps somebody’s used a balloon to make your hair stand on end, or you’ve walked across the office and been ‘shocked’ by a metal door knob? That’s electrostatic discharge at work.