4th Gen Intel Xeon Scalable Sapphire Rapids Leaps Forward


Why Acceleration Matters for Sapphire Rapids

Intel had an observation that was partly self-serving but also valid. SPEC CPU2017 is largely focused on scaling relatively simple tasks across cores and scaling very well. Intel’s observation is that its customers are running VMs, containers, AI, and more. We are at an interesting point where the SPEC CPU2017 is still the standard, but at the same time, whenever we discuss with larger hyper-scalers, they are less excited about the applicability of SPEC CPU2017 to their workloads. That seems to align with Intel’s slide, but also the figures that we found from our SPEC CPU2017 testing.

4th Gen Intel Xeon Scalable Sapphire Rapids SPEC V Workloads
4th Gen Intel Xeon Scalable Sapphire Rapids SPEC V Workloads

We actually did an entire deep dive on the acceleration. We were using pre-production Platinum 8490H’s with all accelerators turned on, but we were not allowed to share the model number at the time. The summary, though is that when we hit accelerators either via on-core instructions or on-package accelerators, performance is very different.

Intel Pre Production Sapphire Rapids Preview QAT Nginx HTTPS Performance Preview
Intel Pre Production Sapphire Rapids Preview QAT Nginx HTTPS Performance Preview

We are going to punt a bit to theΒ Hands-on Benchmarking with Intel Sapphire Rapids Xeon Accelerators piece so you can see the impact of the Sapphire Rapids accelerators across things like the in-core AMX and the QAT accelerators.

Still, we wanted to show why acceleration matters in a use case that was pertinent to us. As a result, we bootstrapped the nginx QAT acceleration to the 32-core 8462Ys, and then ran the full STH nginx stack with the database (and minus back-end tasks like backups/ replication and such) all on a single node and compared it to the AMD EPYC 9374F. Here is what we saw:

Intel V AMD At 32 Cores STH Nginx Stack Performance QAT Impact
Intel V AMD At 32 Cores STH Nginx Stack Performance QAT Impact

QAT is only accelerating a portion of the workload, perhaps 10% but also one that can introduce jitter. At the same time, using the QAT offload changes which solution ends up ahead even though the EPYC 9374F is a 3.85GHz base clock chip with 256MB of L3 cache and the Xeon is a 2.8GHz base clock chip with 60MB of L3 cache.

The reason we highlighted the fact that most 4th Gen Intel Xeon SKUs do not have QAT enabled is that QAT is a path for performance per core gains for Intel in this generation. We will note though, that the QAT setup did require an hour or two of setup, and that was faster since we had the recipe.

4th Gen Intel Xeon Scalable Sapphire Rapids: Power Consumption

In terms of power consumption, the Sapphire Rapids systems were actually better than we expected. Our Supermicro dual socket test platform was fairly consistently hitting peaks of 900-950W using the Intel Xeon Platinum 8490H 60-core parts.

Supermicro SYS 221H TNRR 2U Intel SPR Overview Without Airflow Guide
Supermicro SYS 221H TNRR 2U Intel SPR Overview Without Airflow Guide

We had this system running side-by-side with the ASUS RS720A-E12-RS24U with the 96-core parts. AMD has a slightly higher TDP, but the Intel server used less power. Part of that may be due to the PSUs. Other parts may be due to the cooling configuration and the fact that we had four fewer DIMMs per socket or eight total, which accounts for ~40W + 15% cooling overhead, so around 46W from just the memory difference.

Supermicro SYS 111C NR 1U Intel SPR PSU
Supermicro SYS 111C NR 1U Intel SPR PSU

Our single-socket Supermicro platform was consistently in the 400W range with the slightly lower-end SKUs. While that may seem like a lot, it is actually important. For things like VMware licensing, consolidating two older PCIe Gen3 Skylake era 16-core sockets to a single 32-core processor system saves on license fees. Power consumption is likely equal or lower. One also gets more PCIe bandwidth and lanes and more memory bandwidth than older 2 socket servers. Intel finally has a straightforward single socket consolidation case fitting into 32-core license packs easily while maintaining a similar power footprint.

Next, let us get to our market impact and final words.


  1. Wow … that’s a lot of caveats. Thanks for detailing the issues. Intel could really do with simplifying their SKU stack!

  2. Not sure what to think about power consumption.

    Phoronix has average power consumption reported by sensors that is ridiculously high, but here the peak power plug consumption is slightly less than Genoa.

    Someone needs to test average plug power on comparable systems (e.g. comparable nvme back-end).

  3. This is like BMW selling all cars with heated seats built into them and only enabling it if you pay extra.

    Intel On Demand is a waste of engineering, of silicon, of everything, to please shareholders.

  4. I’ve only made it to the second page but that SKU price list is straight up offensive. It feels like Intel is asking the customer to help offset the costs of their foundry’s missteps for the past four years.

    The segmentation is equally out of control. Was hoping Gelsinger was going to reign it in after Ice Lake but I got my answer loud and clear.

  5. New York Times: “Inside Intel’s Delays in Delivering a Crucial New Microprocessor

    The company grappled with missteps for years while developing a microprocessor code-named Sapphire Rapids. It comes out on Tuesday.”

    – NOT how you want to get free publicity for a new product!

  6. I was so focused on Intel having fewer cores than AMD with only 60 I forgot that there’s still a big market for under 205W TDP CPUs. That’s a good callout STH

  7. Intel did similar things when they lost track versus RISC/AMD back in the day. Itanium, Pentium IV (netburst), MMX and SSE were the answers they used to stay relevant.

    P4’s overheated all the time (think they have this solved today with better cooling, but power is still a heavy draw).

    MMX and SSE were good accelerations, complicating compilers and developers lives, but they existed on every Intel CPU, so you had a guaranteed baseline for all intel chips. Not like this mess of sku’s and lack of predictability. QAT has been around a while, and lots of software support, but the fact it’s not in every CPU holds it back.

    The one accelerator that doesn’t need special software is HBM yet they limit that to too few SKUs and the cost is high on those.

    This is not a win for Intel…this is a mess.

  8. I’ve just finished reading this after 90min.

    THAT is someone who’s got a STRONG understanding of the market. Bravo.

    Where’s the video tho?

  9. There is soomething wrong with the pricing for these products.

    Especially with accelerators there is a price thing going on:
    -QAT can’t compete with DPUs; as you mentioned those cost $300 more than a NIC
    -AMX on $10k+ CPUs (with 56 or 60 cores) can’t compete with a $1500 GPU while consuming much more power than a CPU with less workload plus the GPU.

    These sticker prices might not be end-prices. High core Genoa is also available now ~20% under MSRP from european retailers. I don’t really trust MSRP for this generation.

  10. @Lasertoe – What we’re seeing here is the first step towards the death of the DPU. What is going to be ending it is when Intel integrates networking fabrics on package and thus you can dynamically allocate cores towards DPU tasks. This provides the flexibility, bandwidth and latency that dedicated external cards will quickly disappear.

    Intel isn’t doing themselves a favor by having their on-die accelerators behind the On-Demand paywall.

  11. Hello Patrick
    I suspect you will earn lots of money if you could monetize your Intel SKU excel sheet πŸ™‚
    How on Earth I can pick the best CPU for my workloads ?
    Are there any tools that could identify which accelerations might be helpful for my workloads ?

    Whole concept of the On Demand is kinda rotten.
    I deploy the platform, I migrate the workloads, I realize that maybe some additional accel will be beneficial (how ?), I purchase the extra feature (and it won’t be cheaper if purchased from the get go), and then I need to trigger workload wide software refresh into acceleration enabled version ?
    Hard to see that.
    Sorry if the accelerators are meant to be decision factors there need to be widely adopted, they need to be a must, a no brainer. And they need to have guaranteed future.

  12. I’m extremely confused how NONE of the “Max” SKUs are being offered with ANY of the onboard accelerators! (other than DSA, which seems like the least helpful accelerator by far.)

    Is that a typo? The Max SKUs don’t even offer “on demand”?

  13. @Kevin G:

    I don’t think that will happen. I think Intel and AMD will both integrate DPU-like structures into their server CPUs.

    Allocationg cores “towards DPU tasks” is already possible when you have an abundance of cores like Genoa (and even more with bergamo). The DPU advantage is that those (typically ARM) cores are more efficient, don’t need a lot of die area and don’t share many resources with the CPU (like caches and DRAM).

    I can see a future where efficient coress with smaller die area like Zen4c or Atom (or even ARM/RISC-V) work along high-performance cores for DPU tasks but they need independent L3 caches and maybe DRAM.

  14. Well, have to admit, I didn’t think there would be anything below the $1,500 mark. Granted, there’s not much, but a few crumbs. Now to see if you can actually get those SKUs.

    Not buying the power levels until I see some actual test results. Frankly the lack of accelerators on so many of the high end SKUs definitely raises a few doubts as well. Why leave the thing you’ve been hyping up all this time from so many SKUs, and does this mean that there are, 4-5 different chip lines being manufactured? Thought one of the main angles was that they could just make a single line and bin those to make your variations and offer the unlocks to all the models?

    Just waiting for all the “extras” to become a recurring subscription. You want the power efficiency mode turned on? That’s $9.99/hr/core.

  15. “4th Gen Intel Xeon Scalable Sapphire Rapids Leaps Forward in Lead Times” Fixed the title for you πŸ˜‰

  16. Can anyone explain the difference between the Gold 5000 and Gold 6000 series? I can’t find any rhyme or reason to the distinction.

    Adding to the confusion, the Gold 5415+ actually appears to be substantially worse than the Silver 4416+, and the Silver 4416+ costs $110 more. Why would a Silver processor cost more than a Gold processor and be better? There’s a pretty meaningless-looking distinction in base clocks, but given where the all-core turbo is at, I would bet that loading 8 cores on the 4416+ would yield clock speeds that aren’t far off from the all-core turbo clock speed of the 5415+… and then you still have another 12 cores you can choose to load up on the 4416+, with over 50% more cache!

    The SKU matrix doesn’t seem very well considered. I also agree with Patrick’s comments on the confusing state of the accelerators; I think Intel should have enabled 1 of every accelerator on every single SKU, at a minimum. If they still wanted to do “On Demand”, that could allow users to unlock the additional accelerators of each type, but even having 1 would make a significant performance difference in workloads that can use them, and it would be an effective way to draw the customer into buying the licenses for additional accelerators once they are already using them.

  17. Will be interesting to see The hedt platform later how it Will perform campare to rapid lake ryzen and of course threadripper and also IF they have some new things outside of pci-e5 ddr5 or IF they cripple it as they did with x266

  18. What an absolute mess. The naming has been awful since the whole “Scalable” marketing debacle but this is taking it to the next level. Was hoping they would sort it this generation. Sigh.

  19. Accelerators have a chicken vs. egg adoption challenge. Intel hedged their bet with “on demand,” which makes adoption failure a self-fulfilling prophecy

  20. I don’t know if anyone noticed, but in the chart on page 12 where Intel basically denounces the SPEC benchmarks they put “Gaming” twice in the “Customer workloads” set in relation to the release of a Xeon line.

  21. A lot of games require servers for multiplayer gaming, don’t they? Then of course you have cloud gaming, which is much smaller, I’d imagine.
    It does seem odd that they selected two customers with gaming workloads when there aren’t so many total.

  22. “On Demand” is bullshit. It’s nothing more than artificial scarcity, a.k.a the Comcast model. I would be very angry if I paid for all of those transistors and over half of them were locked behind an additional paywall.

  23. Thanks for the nice article. Unfortunately on general purpose computing it seems Intel is still trying to catch AMD and not successfully.
    I’m using phoronix benchmarks geometric means (from set of benchmarks) comparison here with the specified CPU TDP, E.g. benchmark number / TDP = X. so this basically shows efficiency of processing in comparison with declared TDP. Higher number, better efficiency.
    Intel 8280: 1.35
    Intel 8380: 1.46 — looks like 14nm -> 10nm transition was moderately successful
    Intel 8490H: 1.7 — again 10nm -> Intel 7 although it should be basically same, it looks like Intel did their homework and improved quite a lot.
    AMD 9554: 2.3 — and this is from completely different league. TSMC simply rocks and AMD is not even using their most advanced process node.

  24. Not sure if I get it right. It does seems like 8490H and 8468H had all accelerators enabled from the table you compiled

  25. I don’t find these particularly compelling vs. AMDs offerings. The SKU stack is of course super complicated, and the accelerator story doesn’t sound very compelling – also raises the question if one can even use these with virtualization. And I don’t think most software supports the accelerators out of the box with the possible exception of QAT. The on-demand subscription model also bears the risk that Intel might not renew your subscription at some point.

  26. Those SPECint numbers are ******* BRUTAL for Intel. I’m sure that’s really why they’re saying it’s not a good benchmark. If it’d been reversed, Intel will say it’s the best.

  27. I’d agree on the speccpu #’s.

    I read this. It took like 2hrs. I couldn’t decide if you’re an intel shill or being really critical of intel. I then watched the video, and had the same indecision.

    I’d say that means you did well presenting both sides. There was so much garbage out there at least there’s one place taking up the Anandtech mantle.

  28. Amazing review. It’s by far the most balanced on the Internet on these. I’ll add my time, it took me about 1.25 hours over 3 days to get through. That isn’t fast, but it’s like someone sat and thought about the Xeon line and the market and provided context.

    Thx for this.

  29. I think Intel is on the wrong path.
    They should be making lower powered CPU’s.

    Their lowest TDP CPU is 125W and its a measly 8 core, with a 1.9Ghz max boost frequency – I think something is wrong in Intel’s development department.

    1.9Ghz boost frequency should not require 125W TDP.

  30. Patrick’s SKU tables show the 8452Y as MCC, but that’s clearly impossible since it has 36 cores. It should be XCC (which would also match Intel’s table).

    I didn’t try to check all the others. πŸ™‚


Please enter your comment!
Please enter your name here

This site uses Akismet to reduce spam. Learn how your comment data is processed.