Intel Data Center GPU Max Series

GPTKB entity

Predicate	Object
gptkbp:instanceOf	gptkb:graphics_card
gptkbp:alsoKnownAs	gptkb:Ponte_Vecchio
gptkbp:announced	Intel Innovation 2022
gptkbp:architecture	gptkb:Intel_Xe_HPC
gptkbp:competitor	gptkb:NVIDIA_H100 AMD Instinct MI250
gptkbp:connects	gptkb:EMIB gptkb:Foveros
gptkbp:designedFor	data centers
gptkbp:energyEfficiency	high
gptkbp:firstReleased	2022
gptkbp:formFactor	gptkb:OAM gptkb:PCIe
gptkbp:manufacturer	gptkb:Intel
gptkbp:maximumComputeUnits	128 Xe cores
gptkbp:maximumFP32Performance	104 TFLOPS
gptkbp:maximumFP64Performance	52 TFLOPS
gptkbp:memoryType	gptkb:HBM2e
gptkbp:multiTileDesign	yes
gptkbp:notableModel	gptkb:Intel_Data_Center_GPU_Max_1100 gptkb:Intel_Data_Center_GPU_Max_1350 gptkb:Intel_Data_Center_GPU_Max_1450 gptkb:Intel_Data_Center_GPU_Max_1550
gptkbp:numberOfTiles	up to 47
gptkbp:processNode	gptkb:TSMC_N5 gptkb:Intel_7 TSMC N7
gptkbp:productType	gptkb:Intel_Xe
gptkbp:RAM	128 GB
gptkbp:supports	gptkb:SYCL gptkb:oneAPI gptkb:machine_learning gptkb:PCIe_Gen5 gptkb:CUDA_(via_compatibility_layer) gptkb:OpenCL gptkb:CXL_1.1 deep learning high performance computing artificial intelligence workloads BF16 INT8 FP32 FP64 Tensor operations
gptkbp:targetMarket	cloud computing supercomputing enterprise AI
gptkbp:usedIn	gptkb:Aurora_supercomputer
gptkbp:bfsParent	gptkb:Intel_Ponte_Vecchio gptkb:Aurora_supercomputer gptkb:Intel_AI_Products
gptkbp:bfsLayer	7
https://www.w3.org/2000/01/rdf-schema#label	Intel Data Center GPU Max Series