- Cloud ServicesDeep Learning
- ProductsRackmount &
- Cloud Services
- Deep Learning Servers
- Rackmount Servers
- Blade Servers
- Legacy Products
- SolutionsFor Today's
- IndustriesHPC and more
- CompanyAbout Cirrascale
- ContactGet in Touch
Boost Up to NVIDIA® Tesla® K80 / Accelerate Your GPU Applications
Engineered to Maximize Application Performance
Dual GPU Accelerator
Each NVIDIA® Tesla® K80 has a dual GPU design that allows for higher overall application throughput. Each Cirrascale GB5400 / GB5600 Series server can use eight NVIDIA Tesla K80 cards, enabling up to 16 GPUs to really accelerate your application.
Dynamic GPU boost automatically maximizes application performance by taking advantage of any available power head room.
24GB GPU Memory and 2X Shared Memory
Double memory enables the K80 to run bigger data applications and twice the shared memory enables more concurrent threads to deliver significant speedup without changes to GPU-accelerated code.
Create the Ultimate Multi-GPU Peering Platform
Experience true peer-to-peer enablement over a single root complex when NVIDIA Tesla K80 Dual-GPU Accelerators are installed in the Cirrascale GB Series with its SR3514 PCIe switch riser technology.
Download and Discover
Below are some useful guides provided by NVIDIA and Cirrascale to help you learn more about how to Boost Up with K80s and Boost Your Code.
Each NVIDIA Telsa K80 Accelerator contains TWO GPUs to help boost your code to levels never before experienced.