shopify analytics ecommerce

(+44) 0800 180 4801

Tesla K80 - Top Angle

More Views

NVIDIA Tesla K80 Passive - 24GB GDDR5 GPU Card - 8.74TFLOP SP/2.91TFLOP DP - 2x 2496x GPU Cores - Retail

NVIDIA Tesla K80 has been found to be the fastest accelerator so far in the whole world

Availability: In stock

£4,370.26 Excl. VAT

Key Features

  • 2.91 Tflops Double Precision
  • 8.74 Tflops Single Precision
  • 4992 CUDA Cores
  • 24GB GDDR5 Memory
  • 300W Power Consumption


If you have been looking for high-performance computing, then NVIDIA Tesla K80 could be the best choice for you. This is because it comes with some design innovations which speeds up your GPU-accelerated application as it boosts the performance of your data centre. NVIDIA Tesla K80 has been found to be the fastest accelerator so far in the whole world. Advantageous features of NVIDIA Tesla K80 The Tesla K80 has some powerful features that have made it the most powerful GPU compute card available for Computing Finance, health & research simulation, geological research, airflow dynamics etc... Let us look at those characteristics one by one. 1. It has a dual shared memory Tesla K80 has a dual GPU with a dual memory and a memory bandwidth. This makes the NVIDIA Tesla K80 faster than former Tesla Cards, Tesla K20x and K10 by a very high percentage of 80%. This also makes it outperform the CPU up to ten times on the hundreds of applications. 2. It has a Memory Protection To both the internal memories and the external GDDR5 DRAM of the NVIDIA Tesla K80 there is an Error Correcting Codes (ECC) memory protection. This meets all the requirements for the computing accuracy and the reliability in the super computing and data centres. 3. It has the System Monitoring This manages the GPU processors in the computing systems. It uses HPC Cluster and the grid management solutions in the management of the GPU processors. 4. It has NVIDIA GPU Boost This is for the maximum application performance. The Boost will dynamically scale the clocks depending on the nature of the workload. With this feature of the Tesla K80, you can be assured that each of the applications runs at the highest clocks remaining within the envelope of power and thermal. 5. It has a zero-power idle This is meant to increase the efficiency of the data centre energy. It manages this by powering down the idle GPUS when running legacy workloads that are non-accelerated. 6. It has a Multi-GPU Hyper-Q This feature is mainly to schedule the MPI ranks across the GPU in an efficient and comfortable way. It also increases the utilization of the GPU enhancing its performance. Advantages of NVIDIA Tesla K80 NVIDIA Tesla K80 has a massively parallel CUDA architecture which features many of the must have’ HPC Cluster capabilities. The HPC capabilities here include the ECC (Error Correcting Codes) memory that ensures that there are compromised accuracy and reliability in the performance of the device. The NVIDIA Tesla K80 GPU accelerator also has more than teraflop peak double precision performance. This makes it ideal for most of the dynamic high-performance computing workloads like climate, weather modelling, biochemistry Simulation, Computational Finance, CFD, CAE and computational physics. The other advantage of this device is that it delivers a performance which is equivalent to that of CPU but only a system at a tenth and a twentieth the power consumption. This makes it even better in performance than the CPU in the computing services.

GPU Architecture: 2 × GK210
NVIDIA® CUDA® Cores: 4992
Processor Core Clock: 560MHz
Processor Boost Clock: 562 - 875 MHz
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)
Capacity: 24GB GDDR5 (12 GB per GPU)
Memory Interface: 384-bit
Memory Bandwidth: 480 GB/Sec (Per Board) & 240 GB/s (per GPU)
Memory Clock: 2.5GHz
Cache: L1 & L2
Bus Type: PCI-Express 3.0 x16
SMX: Yes
Dynamic Parallelism: Yes
Hyper-Q: Yes
Thermatic Solution: Passive
235 W
Energy Star Enabling: Yes
1 × 8-pin CPU power connectors
(2 x 8-pin PCIe to single 8-pin CPU adapter included)

Customer Reviews

Write Your Own Review

You're reviewing: NVIDIA Tesla K80 Passive - 24GB GDDR5 GPU Card - 8.74TFLOP SP/2.91TFLOP DP - 2x 2496x GPU Cores - Retail

How do you rate this product? *

  1 star 2 stars 3 stars 4 stars 5 stars
Build Quality
Turnaround Time
Likely To Recommend
Overall Experience

NVIDIA Tesla K80 Passive - 24GB GDDR5 GPU Card - 8.74TFLOP SP/2.91TFLOP DP - 2x 2496x GPU Cores - Retail