Statements (57)
| Predicate | Object |
|---|---|
| gptkbp:instance_of |
gptkb:Graphics_Processing_Unit
|
| gptkbp:architecture |
gptkb:GCN
|
| gptkbp:competes_with |
gptkb:NVIDIA_Tesla
|
| gptkbp:features |
gptkb:Reliability
Scalability Low latency Multi-GPU support Virtualization support Vulkan support Security features High throughput High bandwidth memory Dynamic power management FP16 support Asynchronous compute Error-correcting code memory Direct X support Open GL support Compute shader support FP64 support |
| gptkbp:focus |
gptkb:Artificial_Intelligence
Deep learning Machine learning |
| gptkbp:form_factor |
gptkb:PCIe
|
| gptkbp:has_units |
gptkb:64
|
| https://www.w3.org/2000/01/rdf-schema#label |
Radeon Instinct
|
| gptkbp:manufacturer |
gptkb:AMD
|
| gptkbp:memory_type |
gptkb:HBM2
|
| gptkbp:performance |
High performance computing
|
| gptkbp:power_consumption |
300 W
|
| gptkbp:release_year |
gptkb:2017
|
| gptkbp:successor |
gptkb:Radeon_MI_series
|
| gptkbp:supports |
gptkb:ROCm
gptkb:Open_CL Scientific computing Data analytics Big data processing Financial modeling AI workloads Simulation workloads Image processing workloads Rendering workloads Video processing workloads |
| gptkbp:target_market |
Data centers
|
| gptkbp:uses |
gptkb:Tensor_Flow
gptkb:Infinity_Fabric gptkb:Keras gptkb:Caffe gptkb:Chainer gptkb:MXNet gptkb:Py_Torch gptkb:Oni Machine learning frameworks CUDA-compatible frameworks Open CL-based frameworks |
| gptkbp:bfsParent |
gptkb:AMD_Radeon_Instinct_MI25
|
| gptkbp:bfsLayer |
6
|