Statements (65)
| Predicate | Object |
|---|---|
| gptkbp:instanceOf |
computer architecture
|
| gptkbp:developedBy |
NVIDIA
|
| gptkbp:enhances |
image processing
machine learning data analytics natural language processing robotics scientific computing video processing AI workloads |
| gptkbp:features |
enhanced security features
HBM2 memory Tensor Cores advanced error correction advanced power management dynamic parallelism enhanced debugging tools improved cache architecture improved interconnects improved synchronization mechanisms |
| https://www.w3.org/2000/01/rdf-schema#label |
Volta architecture
|
| gptkbp:improves |
energy efficiency
performance per watt graphics rendering algorithm performance AI inference compute capabilities |
| gptkbp:includes |
NVLink
new memory architecture hardware_support_for_AI |
| gptkbp:introduced |
new instruction set
|
| gptkbp:performance |
parallel processing
|
| gptkbp:predecessor |
gptkb:Maxwell_architecture
|
| gptkbp:produces |
12 nm process
|
| gptkbp:provides |
high availability
high throughput better thermal management scalable performance improved reliability high memory bandwidth robust software ecosystem flexible programming model multi-instance_GPU_capability |
| gptkbp:releasedIn |
2017
|
| gptkbp:successor |
Pascal_architecture
|
| gptkbp:supports |
CUDA
big data processing deep learning high-performance computing real-time ray tracing virtual reality edge computing data-intensive applications high-performance graphics large-scale simulations FP16 precision |
| gptkbp:usedIn |
cloud computing
data centers financial modeling AI research autonomous vehicles supercomputers NVIDIA Tesla V100 NVIDIA Titan V healthcare_applications |