It's The End of Road For Intel's Ponte Vecchio GPU, Company Now Focused On Falcon Shores Development

It's The End of Road For Intel's Ponte Vecchio GPU, Company Now Focused On Falcon Shores Development

Intel has ended deployment of its Ponte Vecchio GPU, marking the end of the road for a chip considered a technological marvel.

Intel's Ponte Vecchio GPU was a technological marvel with its chiplet-heavy design but the Blue Team is now moving on!

Intel's The Ponte Vecchio GPU was first revealed in 2019. and was the brainchild of Intel's former GPU guru, Raja Kodori. When it was announced, the chip was designed to power next-generation exascale computing platforms, but the company had to overcome a number of hurdles to create this masterpiece of a chip that has several on a single package. Chaplets were placed. For reference, Intel's Ponte Vecchio GPU had a total of 47 tiles on a package including:

  • 16 Xe HPC (internal/external)
  • 8 Rambo (Internal)
  • 2 Xe base (internal)
  • 11 EMIB (Internal)
  • 2 Xe Link (External)
  • 8 HBM (External)

Ponte Vecchio GPUs saw a home at Intel Data Center Max GPU Series And one of the lead products built on this exascale chip platform was the Aurora supercomputer, which managed to break the exascale barrier, but only a little too late. Not only did the AMD-powered Frontier supercomputer succeed. Beats the Intel-powered Aurora in the Exascale race. But it currently sits in the No. 1 spot and has a higher efficiency than the system running the Ponte Vecchio.

Managed by Intel Break some AI performance records with the Aurora supercomputer. Thanks to its Xe hardware which includes dedicated AI accelerators but the company is now focusing on its Gaudi accelerators with Gaudi 2 and Gaudi 3 which are now the main chips to serve this segment.

Talking to Server The Home, Intel confirmed that it will no longer deploy clusters with Ponte Vecchio GPUs. The company will continue to offer Ponte Vecchio in existing clusters but no new clusters will be built. For those interested in harnessing the HPC capabilities of the Ponte Vecchio GPUs, they will still be available on the Intel Developer Cloud, but the company has confirmed that it will now turn its attention to the next-generation Falcon Shores GPUs. Focusing on

As of last week, Intel Ponte Vecchio is moving into a new phase. Instead of hunting for new clusters, it will continue to sell and fill existing clusters. Likewise, the Intel Xe architecture is important to the company, so Intel will continue to develop the software behind Intel Xe as it moves forward at Falcon Shores, hopefully next year.

By ServerTheHome

Intel was also expected to launch. Rialto Bridgean upgraded version of the Ponte Vecchio GPUs which was later canned.. And there was the Falcon Shows. Originally the x86 would be a combination of CPU and GPU architecture. But it was abandoned in favor of a GPU-only design. This design would have been similar to the AMD MI300A APU accelerator that combines CDNA 3 GPUs with Zen 4 CPU cores.

Based on what we know, Intel Falcon Shores GPU Gaudi and next-generation Xe will take the best of GPU architecture and combine them into one package. The GPU is slated to launch next year but we'll have to wait and see if the product actually ships on time or faces delays like the Ponte Vecchio that came before it.

Share this story.



About the Author

Leave a Reply