Statements (23)
Predicate | Object |
---|---|
gptkbp:instanceOf |
Neural Engine
|
gptkbp:announced |
2019
|
gptkbp:architecture |
custom AI architecture
|
gptkbp:connects |
gptkb:RoCE_(RDMA_over_Converged_Ethernet)
|
gptkbp:countryOfOrigin |
gptkb:Israel
|
gptkbp:designedFor |
deep learning
|
gptkbp:fabricationProcess |
16nm
|
https://www.w3.org/2000/01/rdf-schema#label |
Gaudi AI processor
|
gptkbp:manufacturer |
gptkb:Habana_Labs
|
gptkbp:memoryType |
gptkb:HBM2
|
gptkbp:notableFeature |
scalable architecture
high throughput for AI workloads integrated Ethernet interfaces |
gptkbp:parentCompany |
gptkb:Intel
|
gptkbp:platform |
gptkb:TensorFlow
gptkb:PyTorch |
gptkbp:successor |
gptkb:Gaudi2_AI_processor
|
gptkbp:supports |
training
inference |
gptkbp:targetMarket |
data centers
|
gptkbp:usedIn |
gptkb:Amazon_EC2_DL1_instances
|
gptkbp:bfsParent |
gptkb:Habana_Labs
|
gptkbp:bfsLayer |
5
|