7 ms·
AMD EPYC 97x4 “Bergamo” CPUs: 128 Zen 4c CPU Cores for Servers, Shipping Now
- WithinReason 3y ago360W per CPU is less than 3W per core, not bad.
- usrusr 3y agoCores won't draw much while blocking on a memory access. If the core count vs core utilization trade-off is a tie, the napkin math expectation would be same throughput, same power draw of cores, less power draw of cache for less cache.
- reckless 3y agoIt's fascinating how they managed to reclaim so much die area by dropping the cache and optimising for low (~3GHz) frequency.
- treesciencebot 3y agoI think we will be seeing this more and more (less cache on the die OR stacked cache on top of the main chip instead of in it) since SRAM scaling is now near a halting point [1] (no more improvements) which means the fixed code of cache is going up with every new node. https://semiwiki.com/forum/index.php?threads/tsmc-officially-halts-sram-scaling.17223/ https://semiwiki.com/forum/index.php?threads/tsmc-officially...
- loeg 3y agoIt's mostly the cache area rather than the frequency, right? Cache just takes a huge amount of space on the chip.
- toast0 3y agoFrom what I've read, the L1/L2 cache per core is the same, and the L3 cache per chiplet is the same, but the cores per chiplet is doubled and the overall chiplet size is about the same (it's a little bigger, I think) So, L3 cache didn't get smaller (in area or bytes), there's just less for each core. L1/L2 is relatively small, but they did use techniqued to make it smaller at the expense of performance. I think the big difference really is the reductions in buffers, etc, needed to get a design that scales to the moon. This is likely a major factor for Apple's M series too. Apple computers are never getting the thermal design needed to clock to 5Ghz, so setting the design target much lower means a smaller core, better power efficiency, lower heat, etc. The same thing applies here: you're not running your dense servers at 5Ghz, there's just not enough capability to deliver power and remove heat; so a design with a more realistic target speed can be smaller and more efficient.
- rektide 3y agoSemianalysis seems to indicate the core space itself is 35% smaller. Maybe that's process tweaks related to clock rate? But I don't think we know for absolute certain it really is the same core with the same execution units & same everything. Even though a ton of the listed stats are exactly the same.
- xeeeeeeeeeeenu 3y ago"Consistent x86 ISA" in that slide is clearly a dig at Intel. AVX-512 is still not present in the chips with E-cores.
- SloopJon 3y agoLast I checked, there aren't any Xeons with E-cores. However, you're right: the most direct competitor to Bergamo seems to be Sierra Forest, a chip that will have nothing but E-cores (and no AVX-512).
- adrian_b 3y agoThe Intel "Sierra Forest" Xeon CPUs with E-cores are announced for 2024. Even if they are not branded Xeon, but Atom, there already are a lot of Intel server CPUs with E-cores, e.g. Denverton Refresh, Snow Ridge, Parker Ridge and the recently launched Arizona Beach.
- api 3y agoA chip with hundreds or thousands of E cores and AVX512 may be competitive with GPUs for AI.
- pclmulqdq 3y agoImagine if that chip also had HBM memory and fit into a PCIe slot, or maybe had direct-attached high-bandwidth networking. Now imagine Intel had that product and canceled it in 2017, and you will be living in reality: https://en.wikipedia.org/wiki/Xeon_Phi https://en.wikipedia.org/wiki/Xeon_Phi
- api 3y agoThey’re fools not to revive that or something like it. Not only would something like 1024 cores with AVX512 be competitive with GPUs it would have the added advantage of being waaay more versatile and easier to program.
- beebeepka 3y agoI wonder if Zen 4c is going to trickle down to the consumer market. There are multiple possibilities for CPUs with mixed chiplets but who knows if they really make sense outside the server world. It's exciting
- happycube 3y agoA 24-core Ryzen with an 8C zen4+3d sram and a 16C zen4c chip might be quite interesting. There are probably already a few in AMD's labs ;)
- wmf 3y agoYeah, that would destroy the 14900K.
- mosquitobiten 3y agoWow, do CPUs have that capability now? To actually destroy other CPUs? How does that work? Do they get the "Annihilate.Competition" instructions set in an update when they detect a Intel machine in the network?/s
- salawat 3y agoHave you never heard of the Intel C compiler's cripple AMD function? https://www.agner.org/forum/viewtopic.php?t=6 https://www.agner.org/forum/viewtopic.php?t=6
- throw7890 3y agoThere was a very detailed post with die shots etc on semi analysis a few weeks back https://www.semianalysis.com/p/zen-4c-amds-response-to-hyperscale https://www.semianalysis.com/p/zen-4c-amds-response-to-hyper... In the subscriber only section it was mentioned there will be some Zen5 consumer parts using Zen5c as the equivalent of current Intel E-cores.
- 3y ago
- speed_spread 3y agoSomething tells me that Teams still won't run properly using this.
- ComputerGuru 3y agoThis doesn’t add any more value to the conversation than Reddit’s “but will it run Crysis?” comments.
- Crystalin 3y agoit didn't but it made me smile :)
- speed_spread 3y agoOk then. Steering away from Redditisms, let's make this into a deep reflection upon the state of modern application software that's consuming hardware resources faster than hardware can improve. We have these incredible CPUs yet our machines offer barely the same experience they did twenty years ago and even succeed in getting strictly worse as in Teams case. The newer the software, the more wasteful implementation we get, our performance budget spilled and absorbed by layers upon layers of abstractions. Of all this immense computing power, I can only feel the difference when compiling stuff and other specialized activities.
- COGlory 3y agoI'm extremely curious if this could run Crysis in with software rendering, actually...
- PragmaticPulp 3y agoIs everyone on HN running their computers in some parallel universe where apps are running 10X slower than my computer?
- fbdab103 3y agoNo, but on the other hand, I do have some glimmer of hope that eventually someone at Microsoft will see these kinds of comments and do something to fix that abomination. I have no doubt Microsoft employs many very talented people, but none of them seem to be working on the user facing products which make my corporate laptop run so poorly (Teams or OneDrive).
- loeg 3y agoWe're going to be using a lot of these in a new hardware sku at my employer sooner or later (it's in the works). We software nerds get to figure out the scaling problems on these big CPUs that don't manifest on the smaller 18 and 24 core CPUs we're using today.
- javajosh 3y agoWe'll approach it like a network-on-a-chip: 128 servers with an extremely fast network. Each server will be a process. One process will be something like ingress, another like a message queue and the rest will be workers. Chances are the chip doesn't need or want to be saturated, so you rotate the cores using Linux magic of some kind. Put that on a napkin and give it to an unsuspecting intern.
- loeg 3y agoCertainly the closer you can get to shared-nothing, the better! Some of the difficulty is we have a legacy stack that is not implemented that way. (You also need to account for significant kernel CPU use, like hardware interrupt handlers (e.g., NIC rx) running on some cores.)
- toast0 3y ago> You also need to account for significant kernel CPU use, like hardware interrupt handlers (e.g., NIC rx) running on some cores. In a lot of loads, it doesn't matter that much, because the application uses way more CPU than packet handling... But if it does matter, you really want to get things lined up as much as possible. Each CPU core gets one NIC queue pinned and one application thread pinned and the connections correctly mapped so the fast path never communicates to other cores. I'm not tuned into current NICs though, I don't know if you can get 128 queues on NICs now. If you have two sockets, then you also have the fun of Non-Uniform PCI-E Access... When I was working on this (for a tcp mode HAProxy install), the NICs did 16 queues, and I had dual 12 or 14-core CPUs, so socket 0 got all its cores busy, and socket 1 just got a couple work threads and was mostly idle. Single socket, power of 2 cores is a lot better to optimize, but I had existing hardware to reuse.
- ksec 3y agoWith 2U, 4 Node, Dual Socket, and 128 Core CPU, you get up to 1024 Core or 2048 vCPU per Server Machine. We can now all enjoy our own cloud.
- fbdab103 3y agoWhile melting the wires in your walls. How much juice would that pull when under full load?
- rektide 3y ago3kw
- Groxx 3y ago^ roughly how much a clothes dryer uses, for comparison. A lot of power, but nothing really abnormal. Most rooms in a house won't be wired for it but that's about all. You probably wouldn't want this much thermal output in a normal room anyway, not to mention fan noise.
- fred123 3y agoIn most parts of Europe the standard room socket has >= 3.6 kW
- magicalhippo 3y agoIt could also double as a clothes dryer (put a cabinet with clothing racks on next to the server rack, pipe the hot air into it). Honey, can you queue up a large simulation? Got a full load to dry again.
- uguuo_o 3y agoThe amount of scientific work that could be done with such a machine would be staggering, specially if it can be local. I use 2-10k cores on a daily basis, having a dedicated machine with 1k would make debugging large scale problems much faster.
- satvikpendem 3y agoNow when can we get the same through Threadripper? I'd love to use my desktop as both my main gaming rig as well as my productivity rig.
- giomasce 3y agoI wonder why AMD decided to name chips after Italian cities...
- jerrygenser 3y agoProbably selected a concept that there were a bunch of names from so that the series would be consistent but also which would not result in any trademark issues. Same reason Intel picks random words and "lake" after
- szatkus 3y agoGeographical names aren't copyrighted, so it's a safe choice. Before that they used names of various rivers around the world. Intel does something similar, but they make up fictional names based on real places. Nvidia uses names of famous scientists.