AMD EPYC 7002 Series Rome Delivers a Knockout


AMD EPYC 7002 Platform and Architectural Updates

The AMD EPYC 7002 is socket compatible with previous generation SP3 platforms. One can install an AMD EPYC 7502P, for example, in a first-generation AMD EPYC server. At the same time, one will likely not get full functionality because the platform has changed. AMD has said the Rome socket is the Milan socket. That tells us that the next-generation Milan AMD EPYC 7003 series is a PCIe Gen4 part. It also tells us that investment in this generation will be socket-compatible with the next generation parts. We will have more on the look-ahead at the end of this article.

AMD EPYC 7002 Platform Updates

We are going to start this section looking at the platform because that is how most of our readers will experience the chips. We already previewed the big revel today in Why AMD EPYC Rome 2P Will Have 128-160 PCIe Gen4 Lanes and a Bonus. We are going to discuss this more about how the second generation AMD EPYC 7002 series is creating a more scalable platform than the previous generation EPYC and current Intel Xeon Scalable.

Here is the overview slide of the AMD EPYC 7002 series. Headlining this release are up to 64 cores/ 128 threads per socket, large caches (up to 256MB per socket), on a quarter of the NUMA domains, reasonable power consumption in the 120-225W range, more floating point performance, PCIe Gen4 support, again at 128 lanes per CPU, and faster memory. That was an enormous sentence. You can see the SKU list in the previous section for full specs.

AMD EPYC 7002 Overview
AMD EPYC 7002 Overview

What AMD is doing in this generation is moving DDR4 memory controllers, PCIe Gen4 I/O, cross-socket interconnects, and even the SATA lanes to a centralized 14nm I/O die. Beyond that, there can be up to 8x 7nm chiplets with the x86 compute that are connected directly to the I/O die.

AMD EPYC 7002 Chiplet Evolution
AMD EPYC 7002 Chiplet Evolution

Key to this is that AMD is presenting a more modern platform architecture. AMD is designing the I/O die as the center of the server. The x86 compute cores hang off of this die like PCIe Gen4 devices, and system memory. In contrast, Intel looks as the x86 die as the center of a computer. We know the future is going to lead to disaggregation with features like CXL interconnects where direct memory access models will work across a system. AMD is already shipping an architecture that follows that future design principle, even if it is not supporting CXL in this generation. AMD’s design is between modern and forward-looking. Intel’s current Xeon lineup is what we will look back on in five years and tell stories that sound like “back in the old days we used to have monolithic die processors.”

AMD EPYC 7002 Chiplet Multi Die Evolution
AMD EPYC 7002 Chiplet Multi-Die Evolution

AMD is being very aggressive because by being an early mover to the modern system architecture, it has a massive advantage in scaling up in a socket and system over Intel.

AMD EPYC 7002 V 2nd Generation Intel Xeon Scalable Comparison
AMD EPYC 7002 V 2nd Generation Intel Xeon Scalable Comparison

We wanted to note here, Intel is moving to chiplet. You can read about Intel Foveros as the direction Intel is moving. AMD is just there in this segment quarters ahead of Intel.

Moving to the I/O die changes the NUMA design of the server. The AMD EPYC 7002 series is presented, by default, as a single NUMA node per socket, down from four in the previous generation. We are going to show tuning this in our later topology section. That change means that there is only 104ns for the same socket latency and 201ns socket-to-socket. This is a 15-19% improvement over the previous generation.

AMD EPYC 7002 Architecture NUMA Reduction To 104ns Close 201ns Far
AMD EPYC 7002 Architecture NUMA Reduction To 104ns Close 201ns Far

Perhaps the biggest change is the I/O subsystem. The diagram will look similar to what we saw in the AMD EPYC 7001 series. In this generation, AMD has PCIe Gen4 giving a 2x per lane improvement over Intel Xeon. AMD also has significantly more lanes. AMD gets up to 128 lanes + an additional lane for the BMC in single socket, and 160 lanes + 2 lanes for BMCs in dual-socket platforms. You can read more about the 160 + 2 configuration in Why AMD EPYC Rome 2P Will Have 128-160 PCIe Gen4 Lanes and a Bonus.

AMD EPYC 7002 Platform PCIe Gen4
AMD EPYC 7002 Platform PCIe Gen4

There is a big catch here. AMD EPYC 7002 platforms need to be a newer spec to take advantage of these features. For example, your first-generation EPYC servers cannot get access to the additional PCIe lane for the BMC by simply adding a second-generation AMD EPYC CPU. Instead, one needs an updated platform. AMD does not broadcast this, but the PCIe Gen4 and DDR4-3200 memory speed support almost always will require a new motherboard PCB to support the higher-speed signaling. We see most of the PCIe Gen4 systems to support the new 2nd Generation AMD EPYC 7002 features.

AMD EPYC 7002 Architecture Memory Speed And Bandwidth Benefits
AMD EPYC 7002 Architecture Memory Speed And Bandwidth Benefits

On the memory side, AMD also leapfrogs Intel Xeon. The 2nd generation Intel Xeon Scalable processors support up to DDR4-2933 when Intel Optane DCPMM is not used and only DDR4-2666 when Optane DCPMM is used. Intel also is based on a six channel design on its mainstream processors.

AMD now has DDR4-3200 memory speed support and eight channels. With 33% more memory channels, and higher speeds, the AMD EPYC 7002 generation has up to 45% more theoretical memory bandwidth. There is more. The AMD EPYC 7002 series also supports up to 4TB of memory per socket. Intel Xeon Scalable 2nd generation only supports up to 1TB on standard SKUs. Intel charges around $3000 more to address up to 2TB per socket (M SKUs) and $7000 more per CPU to go up to 4.5TB (L SKUs.)

At a platform level, this is what the AMD EPYC 7002 looks like compared to the newest Intel Xeon Scalable CPUs:

AMD EPYC 7002 V 2nd Gen Intel Xeon Scalable Top Line Comparison
AMD EPYC 7002 V 2nd Gen Intel Xeon Scalable Top Line Comparison

Here are the AMD EPYC advantages shown on the chart.

AMD EPYC 7002 V 2nd Gen Intel Xeon Scalable Top Line Comparison Chart
AMD EPYC 7002 V 2nd Gen Intel Xeon Scalable Top Line Comparison Chart

When we talk about AMD having a more expansive platform than Intel, and by a large margin, this is exactly why. These are not 10-15% generational improvements. These are more than double in many areas. This is why we are saying AMD has a disruptive platform.

AMD EPYC 7002 Security Platform Updates

AMD has been pushing its security story. Many are familiar with how the company has been less impacted by speculative execution than Intel (e.g. L1TF and Foreshadow.) Beyond that, AMD is building platforms specifically for security. Part of this is an Arm Cortex-A5 security processor within the AMD EPYC I/O die. This provides a hardware root of trust for AMD platforms. We have seen examples of using AMD’s security features already. This includes the HPE ProLiant DL385 Gen10 and DL325 Gen10 that the company says are its most secure servers when combined with iLO5 features. That is the world’s #2 server vendor saying its AMD EPYC platforms are more secure than its higher-volume Intel Xeon servers.

AMD EPYC 7002 Platform Secure Processor
AMD EPYC 7002 Platform Secure Processor

With the first generation of AMD EPYC, the company announced Secure Memory Encryption (SME) and Secure Encrypted Virtualization (SEV). In the 2nd generation AMD EPYC, the company is expanding those capabilities.

AMD EPYC 7002 Platform Secure Memory Encryption
AMD EPYC 7002 Platform Secure Memory Encryption

All memory can be encrypted by a single key which makes attacking memory a harder vector on AMD. In this generation, AMD also can reach to PCIe devices to access encrypted pages.

AMD EPYC 7002 Platform Secure Memory Encryption 2
AMD EPYC 7002 Platform Secure Memory Encryption 2

Another generational improvement example, the AMD EPYC 7002 generation has 509 encrypted memory keys which allow for significantly more guest VMs. In the first generation with AMD EPYC 7001, SEV could handle 15 keys for 15 guests. In this generation, the chip can handle 509 keys which is a much more usable number, especially on systems with up to 128 cores/ 256 threads.

AMD EPYC 7002 Platform Secure Encrypted Virtualization
AMD EPYC 7002 Platform Secure Encrypted Virtualization

In terms of OS Support, most major modern Linux OSes support the feature and we hear that support is coming from VMware and Microsoft.

AMD EPYC 7002 Platform SME SEV Support
AMD EPYC 7002 Platform SME SEV Support

Overall, AMD is pushing its security story because it says customers are asking for more secure platforms. If that is something your virtualization cluster can benefit from and is in your RFP, then AMD has an advantage.

Next, we are going to look at the microarchitecture before seeing what these platform aspects mean for system topology and memory bandwidth.


  1. Absolutely amazing. I still can’t believe the comeback AMD has made in just a few years. From a joke to toppling over the competitor for the top position in what, 3 odd years?

    Definitely going to get this for our next server build. Major props to AMD.

  2. “Intel does not have a competitive product on tap until 2020.”
    Cooper Lake is not remote competitive with Rome, much less it’s actual 2020 competitor Milan.

    Highly unlikely Intel will be close to competitive until it’s Zen equivalent architecture on it’s 7nm node.

  3. Wow! I’ve been holding out upgrading my E5 v3-generation server, workstations, and render farms in my post-production studio because what has been available as upgrades seemed so incremental, it was udnerwhelming. And now here comes Rome and the top SKU is performing 5-6X faster than an E5-2697 v3! Maybe a weird comparison, but specific to me. I’m thinking back to some painfully long renders on recent jobs and imagining those done 5x faster…

    I would really, really love to see some V-Ray or even Cinebench benchmarks. I know I’m not the target market, but I’m not alone in wanting this for media & entertainment rendering and workstation use. Any chance you could run some for us?

    Also, what Rome chip would you need for a 24x NVMe server to make sure the CPU isn’t the bottleneck?

    Great work, as always. Thank you!

  4. Intel’s got Ice Lake too. I’d also wager that Patrick and STH know more about Intel’s roadmap than most.

    Ya’ll did a great job. Using CPU 2017 base instead of peak was good. I thought it was shady of AMD to use peak in their presentations.

    I’d like to see sysbench come back.

  5. Most OEMs will have no problems with moving to Rome but Apple is in a tough situation with their Intel partnership, aren’t they? How can they market Xeon generational improvements when others are will be talking about multiplying performance and a substantial relative price decrease?

  6. Take a look at the top of dual socket systems in the SPECrate2017_int_base benchmark here:
    Supermicro already posted a 655 base with 7742’s to top the charts.

  7. Wizard W0wy – we applied patches, however:
    1. We left Hyper-threading on. I know some have a harder-line stance on if they consider HT on a fully-mitigated setup.
    2. We did not patch for SWAPGSAttack. AMD says they are already patched or not vulnerable here. Realistically, SWAPGSAttack came out the day before our review and there was no way to re-run everything in a day.

    Tyler Hawes – we have the Gigabyte R272-Z32 shown on the topology page. That will handle 24x U.2 NVMe but that will be a common 2U form factor in this generation. CPU selection will depend on NIC used, software stack, and etc., but that is a good place to investigate.

  8. Awesome article STH

    I would love to see some more latency test, Naples had some issues with latency sensitive workloads in part due to the chiplet design. So, will you guys test it out in the future?

    And more database tests?

  9. You did mention you would talk more about 3rd Gen EPYC? I don’t think I saw it anywhere in the article. Will it be out to compete with Ice Lake? What are the claims so far?

    Thanks for the great article! Best I’ve read so far.

  10. I’m also disappointed in the lack of a second gen 7371 SKU. Our aging HP GL380p G8 MSSQL server is due for a replacement, and I don’t want to have to license any more cores. Per-core performance really shines considering $7k/core. It would feel wrong to deploy without PCIe Gen 4; I might drop a 7371 into one of the new boards (if I can get any vendor support) and swap it when the time comes.

  11. I appreciate the amount of work you have done in compiling all this information. Thank you, and well done.

    Also, well done to AMD! What an amazing product they have delivered. Truly one of the greatest leaps in performance-per-dollar we have seen in recent years.

  12. Hello Patrick,
    There was a Gigabyte converged motherboard layout (H262-Z66) floating out that showed 4 Gen-Z 4C slots coming from the CPU. There were rumors of Gen -Z in Rome going back to the Summer of 2018; Is there anything you can tell us about that?

  13. Hi guys, taking my wife to the hospital in 30 minutes for surgery. Will try to get a few more answered but apologies for the delay later today. She broke her elbow (badly.) Thank you for the kind comments.

    Jesper – it is a bit different in this generation. When you are consolidating multiple sockets, or multiple servers, into a single socket, your latency comparison point becomes different as well. We have data but tried to manage scope for the initial review. We will have more coming.

    Luke – Milan is coming, design complete, 7nm+ and the same socket. AMD said the Rome socket is the Milan socket.

    Billy – I think AMD’s problem is that there is so much demand for their current stack, some of those SKUs did not make the launch. I am strongly implying something here.

    Michael Benjamins – 2P 7742 was 27005 without doing thread pinning. There is a lot more performance there. Also, Microsoft Windows Server 2019 needed a patch (being mainlined now) to get 256 threads to boot. I am not sure if I want to show this before we get a better tuned result. Even with this, R20 hits black screen to fully rendered in ~12 seconds. Cores were under 40-98% load for <10 seconds with R20. I actually think R20 needs a bigger test for a 256 thread system.

  14. I’m not sure I understand the paragraph about Intel putting pressure on OEMs. What exactly should not be named/disclosed? Can someone please explain the meaning to me?

    Sounds like the typical and shady anti competitive measures Intel is known for.

    p.s. I hope this is not a double post, but I got no indication if my previous submit worked or not.

  15. Quick question on the successor to Snowy Owl? Have we got an ETA, or will AMD simply pop Ryzen in its place, like ASRock have done?

  16. This is f@#$ing great work. You’ve covered high-level, deep technical, business and market impact, with numbers and practical examples like your load gen servers that are great. I’ve read a few of the other big sites but you’re now on a different level.

  17. To anyone that’s new I’ll reiterate what I said on the jellyfish-fryer article

    Patrick’s the Server Jesus these days.

    He’s done all the server releases and they’re reviewing all the servers

  18. Okay. My criticism was this looked really long. I started reading yesterday. Finished today. Why’d AMD have to launch so late????

    After I was done reading I was totally onboard with your format. You’ve got a lot of context interjected. I’d say this isn’t as sterile as a white paper, but it’s ultra valuable.

    Now get to your reviews on CPUs and servers.

  19. @Youri and another Epyc system from Gigabyte already beat the SuperMicro one at your link ;)

    R282-Z90 (AMD EPYC 7742, 2.25GHz)

  20. I’m thinking you should submit this to some third tier school and call it a doctoral thesis for a PhD. That was a dense long read. I’ve been reading STH since Haswell and I’ll say that I really like how you’ve moved away from ultra clinical to giving more anecdotes. I can tell the difference reading STH over other pubs. This is deep and thorough.

  21. What vendor can accept the first orders for the systems with AMD EPYC 7002 (configurator ready) and is able to ship let’s say within next 2-3 weeks?

  22. I am so glad I waited until today to read this, when I could sit down and read at my leisure. Thank you Patrick and team. This is why I read STH.

  23. “2. Customers to change behavior”

    This is likely not what AMD can do since there is no medicine or medical operation available to fix stupidity!

    Stupidity can’t be fixed by others except people themselves!

  24. Mike Palantir,
    During the event, I thought I recalled the HP rep stating they had systems available for order today.

  25. FYI Rumour rag, WCCF claimed to Fact check your statistic’s!

    “Warning: some of the numbers below are simply absurd.

    ServeTheHome reviewed the top-end 64 core dual socket and found that “AMD now has a massive power consumption per core or performance advantage over Intel Xeon, to the tune of 2x or more in many cases.”

    The new EPYC parts have a massive I/O advantage with 300% the memory capacity versus Xeon 33% more memory channels (8 versus 6) and finally 233% more PCIe Gen3 lanes. But what about actual performance?”

  26. This is probably a dumb question but are there any vendors that will be selling individual chips (not systems) within the next quarter or two? And who would the best vendor be?


  27. guys.. remember that both AMD and us as customers do owe TSMC a lot. Without TSMC all this would probably be not possible today.

  28. Never mind my previous comment. Newegg is selling the processors and is already on back order to the end of August for most of the desirable SKU’s.

  29. Patrick thank you for the informative article and all the great work you and your team do. Also would like to thank the STH readers for their article comments and posts in the forums. This is one of very few sites where I actually enjoy reading what other people think and say…

    And thank you for the nudge nudge wink wink information with regards to the 7371 style skus. I have a application that processes in a very serial fashion and it benefits from higher megahertz vs Core quantity, though 16 cores is perfect for the SQL and other tasks on the machine. I’m excited about the new NUMA architecture and I’m looking forward to whatever is next.

    Best wishes and a speedy recovery to your wife!

  30. @Billy
    Epyc 7542 would probably match or beat the 7371 in mosts lightly threaded tasks.
    What can TSMC make that Samsung couldn’t?

  31. Amazing writeup Patrick, once again! Beamr is proud to be a Day 1 application partner as the only company focused on video encoding. As a result of this amazing achievement by AMD, on the Gen 2 EPYC we were demonstrating at the launch event 8Kp60 HDR live HEVC video encoding on a single socket of a 7742.

    And as a result of having 64 high performance cores, because we are heavily optimized for parallel operation, all cores were utilized at 95% or above! Beamr is super excited to have this level of performance available to our first tier OTT streaming customers and large pay TV operators.

    AMD has broken through on so many levels with this new processor generation that I understand why you feel the need to even go deeper with your analysis and review after writing an “epic” 11k word article.

  32. Great look at the next big thing… After it all, I can only ask if with FINALLY a 1 node socket is there any talk of 4P or 8P…
    The thought of 512C\1024T in a 4U is like dreams come true… And if the rumors of SMT4 turn out to be true (EUV does give 20% more density and power-savings) 512C/2048T could do most heavy jobs in one box…
    And it does change the landscape since the progression from 8C to 64C covers basically 100% of the market.. The market doesn’t care if they need 1P or 8P, they only care about the areas where AMD is excelling…
    Another interesting area I’m not seeing a lot of is Edge Computing… This should seal the deal with an 8 or 16C that can have 6 NICs and an Instinct for AI inferencing…

    Love the site… Looking at bare metal in the future…

  33. So what they’ve figured out that other sites haven’t yet, is the whole consolidation story. That 4 Xeon E5-2630 V4 to 1 epyc really resonates.

  34. It will be interesting to see how long it is before VMware and other companies start adjusting their licensing to reflect future market trends. Software companies have investors to please too. If Intel doesn’t have anything to compete by the time prices start going up then it could cause a huge wave of companies switching to AMD for the simple fact that their licensing would be too expensive otherwise. The other thing they could do is switch. Everything to per core licensing which would give Intel a slight advantage or possibly just a tie once you factor in the total cost. I bet you big changes are coming though. No company could survive having their revenue cut to 1/6 its original value in a couple of years.

  35. So this is me just thinking about this some more. It will also be interesting to see the impact this could have on interest in open source alternatives. Costs jumping 2-6x are the kind of events that get people to start looking into alternatives.

  36. Colby, vmware changing its licensing to per core after appearing and praising rome on stage together with amd, would be one of the top3 stupidest move this industry has ever seen. Not impossible, but highly improbable.

  37. Yeah but in my experience when it comes to looking like an idiot and having to explain to your investors and wall street analysts why your revenue stream has been cut in half most CEOs would prefer to look like an idiot. After all the CEO owns a good portion of the company as well. I don’t necessarily think it will be all at once but instead of a 3% annual increase we may start seeing 10-15%. They also may be hoping that due to the cost reduction allowed in Rome that they will see more customers coming in looking to virtualize since it will be cheaper. Another thing that could potentially go VMware’s way would be if customers just started giving more resources to each vm since they aren’t as constrained by their licensing anymore. Instead of dual core vms with 4GB of ram now everyone gets

  38. …everyone gets 4 cores and 8GB with the benefit to the company being added productivity. Nothing happens in a vacuum in business but the question is what factors are going to prevail the most.

  39. Just joining in for the thanks. The most thorough and in-depth review on the net I’ve found so far.

    Also, Patrick, I wish your wife quick and full recovery. So you can get back to benchmarking, that is ;)

  40. I’m surprised at how inexpensive the lower core count 1P processors are. Are these practical in a high end CFD workstation or for other compute intensive workstations ?

    Someone needs to compare the Ryzen 9 3950X ($750) with the soon to be released 16 core Zen2 with the 7302P ($825). Can’t believe a 16 bit Rome EPYC is only $75 more than the R9 ! The 16 core Zen 2 has to be priced between these 2 devices, maybe $800 ?

    With the 7502P (32 cores) selling for $2300, I guess we know the upper end of the price on the Zen2 32 bit Threadripper.

    Another thing to keep in mind is that Zen3 products will be shipping in 15 months or so. They will surely push down the price/performance curve even further. Zen 3 will be 7nm EUV, which should be 20% higher density, lower power consumption and faster clock speeds. Zen 3 Ryzen should be 32 core, TR should be 128 core, EPYC should be 128 or even 256 core !

  41. @Nobody I’m also really curious about the suitability of these chips for a workstation and how they compare to threadripper. Patrick thought the clock speeds on gen 1 EPYC chips were too slow before the 7371 was released.

  42. Devastating. Adding the fact that second generation is compatible to SP3 and vendors have v2-enabled BIOSes out there already is a serious hit. Good job, AMD

  43. Followed the link back from your article on the 7 and 10nm Intel woes. When you wrote this, you expected Intel to be competitive in 2020. Instead Intel’s process woes have messed up the other parts of the company, and they are considering contracting out CPU and GPU production!

    I never thought I’d see Intel mess up so badly on process, and I know I’m not alone on this. It has given AMD a really big doorway, and curiously enough also seems to have opened the doorway further for ARM vendors, due I think to AMD being limited in production capabilities.


Please enter your comment!
Please enter your name here

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