7nm Intel Xe GPU-ovi kodnog naziva 'Ponte Vecchio'



Intel's first Xe GPU built on the company's 7 nm silicon fabrication process will be codenamed 'Ponte Vecchio,' according to a VideoCardz report. These are not gaming GPUs, but rather compute accelerators designed for exascale computing, which leverage the company's CXL (Compute Express Link) interconnect that has bandwidth comparable to PCIe gen 4.0, but with scalability features slated to come out with future generations of PCIe. Intel is preparing its first enterprise compute platform featuring these accelerators codenamed 'Project Aurora,' in which the company will exert end-to-end control over not just the hardware stack, but also the software.

'Project Aurora' kombinira do šest 'Ponte Vecchio' Xe akceleratora s do dva Xeonova višejezgrena procesora zasnovana na 7 nm 'Sapphire Rapids' mikroarhitekturi i OneAPI, objedinjujući API koji omogućuje jedinstvenu vrstu strojne adrese CPU i GPU. Budući da Intel posjeduje računalnu arhitekturu x86, vjerojatno će Xe GPU-ovi, između ostalog, imati i mogućnost obrade x86 uputa. API će moći gurnuti skalarna radna opterećenja na CPU, a GPU-ove skalarne jedinice i vektorska opterećenja na GPD-ove SIMD jedinice optimizirane vektorima. Intelov glavni korak na tržištu računa mogao bi biti znatno smanjeni troškovi softvera zbog API-ja i objedinjavanja strojnog koda između CPU-a i GPU-a. Ljubaznošću slike: Jan Drewes
Source: VideoCardz