EPYC SAP-SD 2-tier benchmark results - EPYC2 is in class of its own

By: anon.1 (abc.delete@this.def.com), August 9, 2019 4:30 pm
Room: Moderated Discussions
Alberto (git.delete@this.git.it) on August 9, 2019 2:12 am wrote:
> Maynard Handley (name99.delete@this.name99.org) on August 8, 2019 8:58 pm wrote:
> > Alberto (git.delete@this.git.it) on August 8, 2019 3:42 pm wrote:
> > > Michael S (already5chosen.delete@this.yahoo.com) on August 8, 2019 8:29 am wrote:
> > > > SPECpower_ssj2008 Results:
> > > >
> > > > Lenovo Think System SR655, AMD EPYC 7742 2.25Ghz - 19,149 Overall ssj_ops/watt
> > > >
> > > > And several more results in the same range.
> > > > https://www.spec.org/power_ssj2008/results/res2019q3
> > > >
> > > > The best non-EPYC2 result:
> > > > ASUSTeK RS720-E9-RS8, Intel Xeon Pt 8280L, 2.7 GHz - 14,274 Overall ssj_ops/watt
> > > >
> > > >
> > > > The best EPYC1 result:
> > > > Dell PowerEdge R7425, AMD EPYC 7601, 2.20 GHz - 11,867 Overall ssj_ops/watt
> > > >
> > >
> > >
> > > Impressive results.
> > > Look like they chosen the best 7nm silicon available at TSMC to assemble these SKUs. Basically they
> > > have two times the cores (same clock) with only 40W more power consumption over Naples, not to mention
> > > the Zen 2 core is stronger so it draw more power than plain Zen on the same process and the
> > > level of power hungry off die interconnection is INSANE.
> > >
> > > A great showcase, now AMD has to prove to be able to supply the channels with Epyc. One thing
> > > is to supply limited quantities, another one millions and millions of SKUs.
> > >
> > > Bet an hypothetical very popular cpu ala 7742 could be rated between 280-300W
> > > to have enough silicon for customers scoring an acceptable profit.
> > >
> > > Great show :). Now i want to see the manufacturing/financial side of the story.
> > >
> >
> > Where do you imagine the limitations will be?
> > (a) We know that TSMC 7nm low power process can supply Apple volumes (~200M/year). Of course 7nm high
> > performance is not EXACTLY the same process, but what reason is there to imagine that a machine tuned
> > to deliver 100s of millions per year, and with a year of experience, can't make millions?
> >
> > (b) Why do you imagine yield to be an issue? Obviously Apple yields are acceptable.
> > And sure, AMD could, in some crazy fantasy, be cherry picking the 5% best performing
> > chiplets and tossing the rest. But how does that make ANY business sense?
> > We've seen the spread of AMD's offerings (eg here:
> > https://www.anandtech.com/show/14694/amd-rome-epyc-2nd-gen/4
> > )
> > and one feature that is remarkable is how TIGHTLY clustered the peak frequencies are,
> > and the regular pattern of the base frequencies relative to core count. It looks like
> > the process is delivering what are actually extremely NON-VARIABLE chiplets...
> >
> > If your theory made any sense, at the very least AMD would
> > be selling the golden, 5% chiplets at a substantial
> > premium, while offering a second tier of average chiplets at rather lower frequency and cost.
> >
> > (c) The third gating factor COULD be assembly of the final product. I've seen nothing yet
> > about the packaging of Rome -- the technology or who does the work. But it doesn't seem to
> > be any particular stretch in terms of grossly higher frequencies, pin/trace densities, or
> > any other important metric. Which suggests that it's business as usual for the packaging,
> > and that there's no reason to believe that packaging will limit the numbers AMD can ship.
> >
> > Which leads me to ask, once again, exactly what it is that you imagine will prevent AMD from being
> > able to ship at the very least the same sort of volumes they've shipped over the past year?
> > (Of course if they were too pessimistic in their forecasts, they may have initial
> > temporary glitches meeting demand, simply because, as they said, they didn't expect
> > that their competition over the next year would be so inadequate...)
> No they will ship, and very likely even larger quantities than past years.
> My point is that this SKU is in class of its own from manufacturing point of view.
> It is thinked for low/medium volume, not certain for destroy Intel.
> After all 7nm process is not a marvel and even on phone SOCs has clearly showed that the
> TSMC claim of half power versus old generation is only a dream for low clocked FPGAs.
> Magically here we have nearly half power, a clear indication that this line is
> a showcase, and definitively not a frontal attack to Intel market share. A more
> relaxed TDP could have signified a declaration of war against the Blue Team.
> Good for Intel, likely AMD do not want a bloody price war against a so bigger competitor.

Naples has the same interconnect. It's true that you don't need to traverse the interconnect if you are NUMA-aware, but it may not be the case all the time. A higher IPC core burns more power but also gets more work done. The statistic presented is effectively ops/watt, and as long as the performance/power ratio (process neutral) is larger than 1 (which architects should ensure), this should not be a surprise at all. Moreover, a larger LLC ensures lower activity on the interconnect (apart from IPC boost), so in AMD's case, doubling the L3 cache likely helps power a lot more than say intel.

Also, afaik, the IOD is clocked at a lower rate than the cores. I think what you have a hard time believing is how this architecture with all its "obvious" flaws could be more efficient than intel.
< Previous Post in ThreadNext Post in Thread >
TopicPosted ByDate
EPYC SAP-SD 2-tier benchmark resultsMichael S2019/08/08 04:18 AM
  EPYC SAP-SD 2-tier benchmark results - EPYC2 is in class of its ownMichael S2019/08/08 07:29 AM
    Wrong headline. Should be "SPECpower_ssj2008 - EPYC2 is in class of its own" (NT)Michael S2019/08/08 07:51 AM
    EPYC SAP-SD 2-tier benchmark results - EPYC2 is in class of its ownAlberto2019/08/08 02:42 PM
      EPYC SAP-SD 2-tier benchmark results - EPYC2 is in class of its ownblue2019/08/08 06:35 PM
      EPYC SAP-SD 2-tier benchmark results - EPYC2 is in class of its ownMaynard Handley2019/08/08 07:58 PM
        EPYC SAP-SD 2-tier benchmark results - EPYC2 is in class of its ownanon2019/08/09 01:00 AM
          EPYC SAP-SD 2-tier benchmark results - EPYC2 is in class of its ownAlberto2019/08/09 02:09 AM
            EPYC SAP-SD 2-tier benchmark results - EPYC2 is in class of its ownblue2019/08/09 07:47 AM
        EPYC SAP-SD 2-tier benchmark results - EPYC2 is in class of its ownAlberto2019/08/09 01:12 AM
          EPYC SAP-SD 2-tier benchmark results - EPYC2 is in class of its ownMaynard Handley2019/08/09 08:25 AM
            EPYC SAP-SD 2-tier benchmark results - EPYC2 is in class of its ownDoug S2019/08/09 08:33 AM
              EPYC SAP-SD 2-tier benchmark results - EPYC2 is in class of its ownanon2019/08/09 09:24 AM
            EPYC SAP-SD 2-tier benchmark results - EPYC2 is in class of its ownAlberto2019/08/10 03:04 AM
          EPYC SAP-SD 2-tier benchmark results - EPYC2 is in class of its ownanon.12019/08/09 04:30 PM
        EPYC SAP-SD 2-tier benchmark results - EPYC2 is in class of its ownme2019/08/09 08:58 PM
      EPYC SAP-SD 2-tier benchmark results - EPYC2 is in class of its ownanon2019/08/08 11:47 PM
  EPYC SAP-SD 2-tier benchmark resultsLinus Torvalds2019/08/09 02:29 PM
    EPYC SAP-SD 2-tier benchmark resultsanon2019/08/10 06:20 AM
      EPYC SAP-SD 2-tier benchmark resultsme2019/08/10 06:36 AM
        EPYC SAP-SD 2-tier benchmark resultsanon2019/08/10 07:20 AM
          EPYC SAP-SD 2-tier benchmark resultsanon.12019/08/10 08:40 AM
            EPYC SAP-SD 2-tier benchmark resultsanon2019/08/10 09:38 AM
          EPYC SAP-SD 2-tier benchmark resultsblue2019/08/10 08:47 AM
        EPYC SAP-SD 2-tier benchmark resultsAdrian2019/08/10 09:43 AM
          EPYC SAP-SD 2-tier benchmark resultsme2019/08/10 10:06 AM
            EPYC SAP-SD 2-tier benchmark resultsMichael S2019/08/10 10:28 AM
    Big gotchas: poor specsjohn2019/08/10 09:54 AM
      Big gotchas: poor specsme2019/08/10 10:04 AM
      Big gotchas: poor specsanon2019/08/10 10:25 AM
        Big gotchas: poor specsLinus Torvalds2019/08/10 11:13 AM
          Big gotchas: poor specsanon2019/08/10 12:58 PM
          Big gotchas: poor caches2019/08/12 12:51 AM
            Big gotchas: poor caches2019/08/12 12:59 AM
            Big gotchas: poor cachesTravis Downs2019/08/12 08:15 AM
              parallelism vs. efficiencyhobold2019/08/13 03:43 AM
            Big gotchas: poor cachesAndy N2019/08/12 05:45 PM
              Big gotchas: poor cachesanon2019/08/12 11:43 PM
          Big gotchas: poor specsJouni Osmala2019/08/12 09:18 PM
            Big gotchas: poor specsAdrian2019/08/12 10:58 PM
              Kernel compile 25-30s, noise 25-27dBJouni Osmala2019/08/13 03:00 AM
                Quiet and fast workstationsGian-Carlo Pascutto2019/08/13 05:00 AM
                  Quiet and fast workstationsAdrian2019/08/13 06:01 AM
                  Quiet and fast workstationsDavid Hess2019/08/13 12:51 PM
                    Quiet and fast workstationsGian-Carlo Pascutto2019/08/13 11:43 PM
                      Quiet and fast workstationsMichael S2019/08/14 01:09 AM
                        Quiet and fast workstationsGian-Carlo Pascutto2019/08/14 01:59 AM
                          Quiet and fast workstationsDavid Hess2019/08/14 05:05 PM
                        Quiet and fast workstationsDavid Hess2019/08/14 04:56 PM
                      Quiet and fast workstationsDavid Hess2019/08/14 05:30 PM
                  Passively cooled threadripper/xeon WJouni Osmala2019/08/15 02:12 AM
                    Passively cooled threadripper/xeon WAdrian2019/08/15 02:47 AM
                      Passively cooled threadripper/xeon WJouni Osmala2019/08/15 04:58 AM
                        Passively cooled threadripper/xeon WAdrian2019/08/15 11:12 PM
                          Passively cooled threadripper/xeon WAdrian2019/08/15 11:21 PM
                          Passively cooled threadripper/xeon WJouni osmala2019/08/16 08:46 AM
                            Passively cooled threadripper/xeon WGroo2019/08/16 11:07 AM
                Kernel compile 25-30s, noise 25-27dBAdrian2019/08/13 05:39 AM
                Kernel compile 25-30s, noise 25-27dBLinus Torvalds2019/08/13 11:12 AM
                  which graphics card?anonymous22019/08/13 11:52 AM
                    which graphics card?Linus Torvalds2019/08/13 12:12 PM
                  Kernel compile 25-30s, noise 25-27dBDoug S2019/08/13 02:22 PM
                    Kernel compile 25-30s, noise 25-27dBDavid Hess2019/08/14 05:38 PM
                  Kernel compile 25-30s, noise 25-27dBJouni Osmala2019/08/13 11:09 PM
                    AIOsGroo2019/08/14 06:19 AM
                      AIOsGian-Carlo Pascutto2019/08/14 07:52 AM
                        AIOsGroo2019/08/14 08:49 AM
                      AIOsBrett2019/08/14 01:02 PM
                    Kernel compile 25-30s, noise 25-27dBLinus Torvalds2019/08/14 10:10 AM
                  Kernel compile 25-30s, noise 25-27dBGian-Carlo Pascutto2019/08/14 12:33 AM
                    Kernel compile 25-30s, noise 25-27dBAdrian2019/08/14 04:06 AM
                      Kernel compile 25-30s, noise 25-27dBGian-Carlo Pascutto2019/08/14 08:01 AM
                      AMD Bios, etc..Anon2019/08/14 10:39 PM
                        AMD Bios, etc..hobold2019/08/15 06:10 AM
                          AMD Bios, etc..Anon2019/08/15 03:20 PM
                        AMD Bios, etc..Jan Olšan2019/08/15 07:49 AM
                          AMD Bios, etc..Adrian2019/08/15 08:16 AM
                          AMD Bios, etc..Gro2019/08/15 09:57 AM
                        AMD Bios, etc..David Hess2019/08/16 04:39 AM
      Big gotchas: poor specsAdrian2019/08/10 12:28 PM
      Big gotchas: poor specsDummond D. Slow2019/08/10 12:57 PM
        Big gotchas: poor specsanon22019/08/10 11:33 PM
          Big gotchas: poor specsblue2019/08/10 11:57 PM
          Big gotchas: poor specsAdrian2019/08/11 04:41 AM
            Big gotchas: poor specsaaron spink2019/08/12 08:45 AM
              Big gotchas: poor specsFoo_2019/08/12 10:02 AM
              Big gotchas: poor specsTravis Downs2019/08/12 10:26 AM
                Big gotchas: poor specsGabriele Svelto2019/08/13 08:48 AM
                Big gotchas: poor specsaaron spink2019/08/13 06:23 PM
                  Big gotchas: poor specsDoug S2019/08/13 10:15 PM
                    Big gotchas: poor specsGian-Carlo Pascutto2019/08/14 12:35 AM
                    Big gotchas: poor specsaaron spink2019/08/15 04:16 PM
                      Big gotchas: poor specsJouni Osmala2019/08/15 08:03 PM
                        Big gotchas: poor specsaaron spink2019/08/19 03:05 AM
                          Big gotchas: poor specsMichael S2019/08/19 03:37 AM
                            Big gotchas: poor specsRicardo B2019/08/19 04:40 PM
                              Big gotchas: poor specsAaron Spink2019/08/25 01:29 AM
              Big gotchas: poor specsAdrian2019/08/12 10:31 AM
              Big gotchas: poor specsJukka Larja2019/08/13 06:37 AM
              Big gotchas: poor specsLinus Torvalds2019/08/13 10:31 AM
          Big gotchas: poor specsDummond D. Slow2019/08/11 05:23 AM
      FX-9590 > i7-9700KGian-Carlo Pascutto2019/08/12 09:37 AM
    RDRAND issueanonymou52019/08/10 01:52 PM
      RDRAND issueLinus Torvalds2019/08/10 03:15 PM
        RDRAND issueGionatan Danti2019/08/11 12:35 AM
          RDRAND issueLinus Torvalds2019/08/11 09:43 AM
            RDRAND issueLinus Torvalds2019/08/11 10:05 AM
              RDRAND issue@never_released2019/08/11 10:10 AM
                RDRAND issueLinus Torvalds2019/08/11 12:13 PM
                  Err....Groo2019/08/11 05:47 PM
                    Err.... This is a zero content fud posteye roll2019/08/11 07:48 PM
                    Err....Brendan2019/08/11 07:50 PM
                    Err....Jukka Larja2019/08/12 06:38 AM
                  RDRAND issueJose2019/08/12 01:33 AM
              RDRAND issueGionatan Danti2019/08/11 01:12 PM
            RDRAND issueaaron spink2019/08/12 08:18 AM
      RDRAND issuename2019/08/11 08:35 AM
Reply to this Topic
Body: No Text
How do you spell tangerine? 🍊