At SC21 we got additional details around Intel Sapphire Rapids, the next-generation Intel Xeon platform. We had the opportunity to chat with Intel prior to SC21 and have a bit more market context around the new chips. Intel also is highlighting the Aurora supercomputer at SC21.
Intel Sapphire Rapids Update at SC21
First, Intel has its Sapphire Rapids Xeon and the HBM model. We have covered these previously at STH. Above that in terms of capability and integration, there is Ponte Vecchio Intel’s high-end HPC GPU/ accelerator. As no surprise, Intel expects to continue making new chips.
For 2022 though, the big ones are Sapphire Rapids for the Xeon side and Ponte Vecchio for the GPU.
We have covered Intel Details Sapphire Rapids several times now. We even saw Samsung showing Intel Sapphire Rapids PCIe Gen5 Server at OCP Summit 2021 and a Flex Bodega Bay Intel Xeon Sapphire Rapids System.
Perhaps the big new disclosure is that Intel is using 64GB of HBM2e on Sapphire Rapids. We just saw HBM3 last week in our Wild Hardware of OCP Summit 2021 Speed Run. Given the product timing we get HBM2e as we see on accelerators like the NVIDIA A100 80GB today.
On the Ponte Vecchio side we get 408MB of L2 cache and lots of performance. Intel is productizing 4, 6, and 8 GPU configurations. This is similar to how NVIDIA is selling its Redstone and HGX A100 8x GPU platforms.
Intel also said that Aurora has over 54,000 Ponte Vecchio GPUs and 18,000 Sapphire Rapids processors. You will note this is a 6:1 ratio. That is something we knew about previously.
In the final words a quick recap of a few questions.
Intel is releasing information about its new chips at an excruciatingly slow pace. Still, it is good to get new disclosures.
One of the big questions that I asked Jeff McVeigh VP and GM of HPC at Intel about was about HBM versus non-HBM Sapphire Rapids Xeon mix in the market. As one would expect, we did not give exact pricing, but he did indicate that the HBM2e equipped processors would carry a significant price premium. As such, it seems like Intel is expecting these to be very different segments. This is expected, but we have had some readers assume that Sapphire Rapids’s popular SKUs would be the HBM2e version. It seems like Intel is expecting higher volumes on the non-HBM models. That makes perfect sense, but it is also a good confirmation.
Another item I asked Jeff about was how to view Ponte Vecchio. I teed up the question noting that NVIDIA seems keen to position its data center accelerators as AI accelerators while the AMD Instinct MI200 seems to be focused on HPC. Jeff said that he expects Ponte Vecchio to target both workloads. Time will tell a bit on that one, but the other key on this question is that Jeff noted that he expected PV to be competitive in both realms. My sense is that Intel probably has some sense of what its competition is offering customers, so this may be a big opportunity for Ponte Vecchio.
Now, we just need to get to 2022 so we can see these chips in the wild since their platforms are already being shown off!