NVIDIA K80

GPTKB entity

Statements (51)
Predicate Object
gptkbp:instanceOf gptkb:graphics_card
gptkbp:architecture gptkb:Kepler
gptkbp:boardHeight 4.4 inches
gptkbp:boardLength 10.5 inches
gptkbp:boardWidth dual-slot
gptkbp:boostClock 875 MHz
gptkbp:coolingSystem passive
gptkbp:coreCount 4992
gptkbp:CUDAcoresPerGPU 2496
gptkbp:formFactor dual-slot
gptkbp:GPU gptkb:GK210
gptkbp:hasDisplayOutputs No
https://www.w3.org/2000/01/rdf-schema#label NVIDIA K80
gptkbp:interface gptkb:PCI_Express_3.0
gptkbp:manufacturer gptkb:NVIDIA
gptkbp:maximumSpeed 300 W
gptkbp:memoryBusWidth 384-bit per GPU
480 GB/s
gptkbp:memoryPerGPU 12 GB
gptkbp:memoryType gptkb:GDDR5
gptkbp:numberOfGPUs 2
gptkbp:passiveCooling Yes
gptkbp:peakFP64Performance 2.91 TFLOPS
gptkbp:predecessor gptkb:NVIDIA_Tesla_K40
gptkbp:productType gptkb:Tesla
gptkbp:RAM 24 GB
gptkbp:releaseDate November 2014
gptkbp:singlePrecisionPerformance 8.74 TFLOPS
gptkbp:speed 560 MHz
gptkbp:successor gptkb:NVIDIA_Tesla_P100
gptkbp:supportsCUDA Yes
gptkbp:supportsDirectCompute Yes
gptkbp:supportsECCMemory Yes
gptkbp:supportsGPUVirtualization Yes
gptkbp:supportsNVIDIACUDAComputeCapability 3.7
gptkbp:supportsNVIDIAGPUBoost Yes
gptkbp:supportsNVLink No
gptkbp:supportsOpenCL Yes
gptkbp:supportsPCIeGen3 Yes
gptkbp:supportsSLI No
gptkbp:targetMarket high-performance computing
gptkbp:TDP 300 W
gptkbp:technology 28 nm
gptkbp:transistorCountPerGPU 7.1 billion
gptkbp:usedIn gptkb:machine_learning
data centers
scientific computing
gptkbp:bfsParent gptkb:NVIDIA_Data_Center_GPUs
gptkb:NVIDIA_P100
gptkb:NVIDIA_Data_Center_GPU_family
gptkbp:bfsLayer 8