Cache and Infinity Fabric

If it hasn’t been hammered in already,  the big change in the cache is the L1 instruction cache which has been reduced from 64 KB to 32 KB, but the associativity has increased from 4-way to 8-way. This change enabled AMD to increase the size of the micro-op cache from 2K entry to 4K entry, and AMD felt that this gave a better performance balance with how modern workloads are evolving.

The L1-D cache is still 32KB 8-way, while the L2 cache is still 512KB 8-way. The L3 cache, which is a non-inclusive cache (compared to the L2 inclusive cache), has now doubled in size to 16 MB per core complex, up from 8 MB. AMD manages its L3 by sharing a 16MB block per CCX, rather than enabling access to any L3 from any core.

Because of the increase in size of the L3, latency has increased slightly. L1 is still 4-cycle, L2 is still 12-cycle, but L3 has increased from ~35 cycle to ~40 cycle (this is a characteristic of larger caches, they end up being slightly slower latency; it’s an interesting trade off to measure). AMD has stated that it has increased the size of the queues handling L1 and L2 misses, although hasn’t elaborated as to how big they now are.

Infinity Fabric

With the move to Zen 2, we also move to the second generation of Infinity Fabric. One of the major updates with IF2 is the support of PCIe 4.0, and thus the increase of the bus width from 256-bit to 512-bit.

Overall efficiency of IF2 has improved 27% according to AMD, leading to a lower power per bit. As we move to more IF links in EPYC, this will become very important as data is transferred from chiplet to IO die.

One of the features of IF2 is that the clock has been decoupled from the main DRAM clock. In Zen and Zen+, the IF frequency was coupled to the DRAM frequency, which led to some interesting scenarios where the memory could go a lot faster but the limitations in the IF meant that they were both limited by the lock-step nature of the clock. For Zen 2, AMD has introduced ratios to the IF2, enabling a 1:1 normal ratio or a 2:1 ratio that reduces the IF2 clock in half.

This ratio should automatically come into play around DDR4-3600 or DDR4-3800, but it does mean that IF2 clock does reduce in half, which has a knock on effect with respect to bandwidth. It should be noted that even if the DRAM frequency is high, having a slower IF frequency will likely limit the raw performance gain from that faster memory. AMD recommends keeping the ratio at a 1:1 around DDR4-3600, and instead optimizing sub-timings at that speed.

Integer Units, Load and Store Conclusions: Platform, SoC, Core
POST A COMMENT

217 Comments

View All Comments

  • BMNify - Saturday, June 22, 2019 - link

    the only real world test that matters in the UHD2/8K Rec. 2020/BT.2020 LIVE NHK/bbc broadast of the 2020 Summer Olympics will begin on Friday, 24 July and related video streams is can AMD Zen 2 do it can any pc core do realtime x264/x265/ffmpeg software encoding and x264/x265 compliant decoding (notice how many hw assisted encoders today dont decode to spec as seen when you re-enode them with the latest ffmpeg), how many 8k encodes and what overheads are remaining if any can even do one... Reply
  • stance_changer - Sunday, June 23, 2019 - link

    Does IF use PCI E? I thought it used the wiring in 2p epyc systems, and IIRC PCI E doesn't double the bus width every gen, but I would love to be proven wrong. Reply
  • SlitheryDee - Friday, June 28, 2019 - link

    I've been using intel for a few years now, but I must say I can't describe how much I love what AMD is doing these days. I go where the performance per dollar is generally, so the best complement I can pay them is to say my next upgrade will be based on an AMD chip. Reply
  • SlyNine - Sunday, July 7, 2019 - link

    So, what time exactly do these new cpus launch. I mean. The hour. Reply
  • Dodozoid - Sunday, July 7, 2019 - link

    Yeah, I was also trying to find that information with no success.
    Do the reviewers know already or are they waiting for a release instruction from AMD?
    Reply
  • ilux.merks - Sunday, July 7, 2019 - link

    What nobody is talking about is how are the fixes for meltdown and spectre on these new amd processors? Reply
  • Korguz - Sunday, July 7, 2019 - link

    simple.. they dont exist, from what i have seen.. those issues.....are intels only ... Reply

Log in

Don't have an account? Sign up now