AMD EPYC Milan Review Part 2: Testing 8 to 64 Cores in a Production Platform
by Andrei Frumusanu on June 25, 2021 9:30 AM ESTCompiling Performance / LLVM
As we’re trying to rebuild our server test suite piece by piece – and there’s still a lot of work go ahead to get a good representative “real world” set of workloads, one more highly desired benchmark amongst readers was a more realistic compilation suite. Chrome and LLVM codebases being the most requested, I landed on LLVM as it’s fairly easy to set up and straightforward.
git clone https://github.com/llvm/llvm-project.gitcd llvm-projectgit checkout release/11.xmkdir ./buildcd ..mkdir llvm-project-tmpfssudo mount -t tmpfs -o size=10G,mode=1777 tmpfs ./llvm-project-tmpfscp -r llvm-project/* llvm-project-tmpfscd ./llvm-project-tmpfs/buildcmake -G Ninja \ -DLLVM_ENABLE_PROJECTS="clang;libcxx;libcxxabi;lldb;compiler-rt;lld" \ -DCMAKE_BUILD_TYPE=Release ../llvmtime cmake --build .We’re using the LLVM 11.0.0 release as the build target version, and we’re compiling Clang, libc++abi, LLDB, Compiler-RT and LLD using GCC 10.2 (self-compiled). To avoid any concerns about I/O we’re building things on a ramdisk. We’re measuring the actual build time and don’t include the configuration phase as usually in the real world that doesn’t happen repeatedly.
In compiling workloads, the 7763 and 75F3 also saw a 3-4% increase in performance compared to their initial reviews.
The 16-core 7343 ends up as the worst performing chip in this metric, while the 24-core 7443 still managed to put itself well ahead of the 28-core Xeon 6330.
I’ve omitted the 72F3 from the chart due to its bad results of >17 minutes per socket rescaling the chart too much – obviously compiling is not the use-case for that SKU.
Facebook
Twitter
RSS
58 Comments
View All Comments
DannyH246 - Thursday, July 1, 2021 - link
lol - no need to be subtle about it. www.IntelTech.com has been doing this for years.Qasar - Thursday, July 1, 2021 - link
hilarious, go back to wccftech then dannywhatthe123 - Friday, June 25, 2021 - link
good god man, the review quite literally posts hard numbers of epyc simply thrashing xeon in performance even on a per core basis, and you think they're worried that intel will retaliate against them if they don't say something nice about one corner of a segment of performance?what is it about technology that attracts cultists?
Threska - Saturday, June 26, 2021 - link
There's a reason the PCMasterRace forum exists.msroadkill612 - Sunday, June 27, 2021 - link
"Cultists" - I like it :)So true, & on many levels. I see strangely neurotic behaviour from such an allegedly smart & rational demographic.
As a group, they are prone to be great at rattling off streams of presumably accurate numbers and jargon, but arrive at childishly naive conclusions, & ask the wrong questions based plain wrong premises.
devione - Sunday, June 27, 2021 - link
Jesus Christ man. Grow some fucking balls. If you're going to call out Anandtech for being biased at least be straightforward and frank. No need to write an essay trying to couch and justify and be obtuse about itMakste - Monday, July 5, 2021 - link
LmaoOxford Guy - Friday, June 25, 2021 - link
What are the RAM timings? I don’t see that information in the charts.Andrei Frumusanu - Saturday, June 26, 2021 - link
These are standardised PC4-3200AA-RB2-12 sticks, running at JEDEC timings.https://www.micron.com/products/dram-modules/rdimm...
Oxford Guy - Monday, June 28, 2021 - link
Thank you. And the other systems tested?