Dissecting Intel's EPYC Benchmarks: Performance Through the Lens of Competitive Analysis
by Johan De Gelas & Ian Cutress on November 28, 2017 9:00 AM EST- Posted in
- CPUs
- AMD
- Intel
- Xeon
- Skylake-SP
- Xeon Platinum
- EPYC
- EPYC 7601
HPC Benchmarks
Discussing HPC benchmarks feels always like opening a can of worms to me. Each benchmark requires a thorough understanding of the software and performance can be tuned massively by using the right compiler settings. And to make matters worse: in many cases, these workloads can be run much faster on a GPU or MIC, making CPU benchmarking in some situations irrelevant.
NAMD (NAnoscale Molecular Dynamics) is a molecular dynamics application designed for high-performance simulation of large biomolecular systems. It is rather memory bandwidth limited, as even with the advantage of an AVX-512 binary, the Xeon 8160 does not defeat the AVX2-equipped AMD EPYC 7601.
LAMMPS is classical molecular dynamics code, and an acronym for Large-scale Atomic/Molecular Massively Parallel Simulator. GROMACS (for GROningen MAchine for Chemical Simulations) primarily does simulations for biochemical molecules (bonded interactions). Intel compiled the AMD version with the Intel compiler and AVX2. The Intel machines were running AVX-512 binaries.
For these three tests, the CPU benchmarks results do not really matter. NAMD runs about 8 times faster on an NVIDIA P100. LAMMPS and GROMACS run about 3 times faster on a GPU, and also scale out with multiple GPUs.
Monte Carlo is a numerical method that uses statistical sampling techniques to approximate solutions to quantitative problems. In finance, Monte Carlo algorithms are used to evaluate complex instruments, portfolios, and investments. This is a compute bound, double precision workload that does not run faster on a GPU than on Intel's AVX-512 capable Xeons. In fact, as far as we know the best dual socket Xeons are quite a bit faster than the P100 based Tesla. Some of these tests are also FP latency sensitive.
Black-Scholes is another popular mathematical model used in finance. As this benchmark is also double precision, the dual socket Xeons should be quite competitive compared to GPUs.
So only the Monte Carlo and Black Scholes are really relevant, showing that AVX-512 binaries give the Intel Xeons the edge in a limited number of HPC applications. In most HPC cases, it is probably better to buy a much more affordable CPU and to add a GPU or even a MIC.
The Caveats
Intel drops three big caveats when reporting these numbers, as shown in the bullet points at the bottom of the slide.
Firstly is that these are single node measurements: One 32-core EPYC vs 20/24-core Intel processors. Both of these CPUs, the Gold 6148 and the Platinum 8160, are in the ball-park pricing of the EPYC. This is different to the 8160/8180 numbers that Intel has provided throughout the rest of the benchmarking numbers.
The second is the compiler situation: in each benchmark, Intel used the Intel compiler for Intel CPUs, but compiled the AMD code on GCC, LLVM and the Intel compiler, choosing the best result. Because Intel is going for peak hardware performance, there is no obvious need for Intel to ensure compiler parity here. Compiler choice, as always, can have a substantial effect on a real-world HPC can of worms.
The third caveat is that Intel even admits that in some of these tests, they have different products oriented to these workloads because they offer faster memory. But as we point out on most tests, GPUs also work well here.
Facebook
Twitter
RSS
105 Comments
View All Comments
Topweasel - Tuesday, November 28, 2017 - link
Yeah I want to give them the benefit of the doubt and I have no problem with them posting numbers even as analyzation of Intel in regards to EPYC. But a full page "review" of Intel's Epyc benchmarks as a product is kind of schilly. I mean where is their tests to back up the information? Where are the counterpart test where they test something similar that wasn't handpicked by Intel. How can any company assess the validity of a product based solely off of it's competitors testing of the product?bmf614 - Tuesday, November 28, 2017 - link
If you could actually get ahold of Epyc they would probably review the hardware themselves but as of yet it is a paper launch.supdawgwtfd - Wednesday, November 29, 2017 - link
It's not a paper launch dipshit.They can bearly keep up with orders for large companies.
Ryan Smith - Wednesday, November 29, 2017 - link
To throw some context in here, the purpose of this article isn't to publish Intel's benchmarks. Rather, it's commentary on what has been a very unusual situation.Up until now, neither AMD nor Intel have engaged in any serious Skylake Xeon vs. Zen EPYC technical marketing.
"AMD's technical marketing of the new CPU has been surprisingly absent, as the company not published any real server benchmarks. The only benchmarks published were SPEC CPU and Stream, with AMD preferring for its partners and third parties to promote performance"
This despite the fact that AMD and Intel's server products haven't been competitive like this in nearly a decade. Normally you'd expect there to be case studies flying out left and right, which has not been the case. And it's especially surprising since, as the underdog, AMD needs to claw back lost ground.
Consequently, Intel's own efforts are, to date, the first efforts by a server vendor to do a comprehensive set of benchmarks over a range of use cases. And let's be clear here: this is Intel doing this for Intel's own benefit. Which is why we've already previously reviewed the two CPUs, as have other 3rd party groups.
Still, I think it's very interesting to look at what Intel has chosen to represent, and what their numbers show. Intel has more resources than pretty much everyone else when it comes to competitive analysis, after all. So their choices and where they show themselves falling behind AMD says a lot about the current situation.
TL;DR: We thought this stuff was interesting, especially since neither vendor until now has done a Xeon-SP vs. EPYC comparison. And since we've already done our own independent review (https://www.anandtech.com/show/11544/intel-skylake... ), it gives us a set of data to compare to our own conclusions (and to be clear, this isn't a review nor are we trying to call it one)
CajunArson - Tuesday, November 28, 2017 - link
Yeah, you were so doing your righteous complaints when Anandtech did literally the same thing for AMD when AMD went out and misconfigured Intel boxes to pretend that Epyc was better than it actually was.Oh wait, you weren't.
ddriver - Tuesday, November 28, 2017 - link
The problem is the heavily biased towards intel AT coverage you clog. How could anyone complain about the opposite when AT have never displayed pro-amd bias? I have a problem with bias, and I point it out when I see it. You can bet your ass the moment AT shows unfair bais toward amd I will be there to point it out. But I cannot point it out if it doesn't exist.Hurr Durr - Tuesday, November 28, 2017 - link
He was too busy ordering specific platters for his thousands of HDDs with one hand and screaming in threads about hypetane with the other.lkuzmanov - Tuesday, November 28, 2017 - link
I've frequented the site for what must be over 10 years, but I fully agree this is, at the very least, a terrible idea.bmf614 - Tuesday, November 28, 2017 - link
Toms and many other sites also covered this.wumpus - Wednesday, November 29, 2017 - link
If Intel suddenly feels the need to compete with AMD, that 's news (practically "man bites dog" news judging from the last decade or so).The fact that they have to pick carefully contrived benchmarks to appear superior to AMD is even more telling. Totally ignoring power consumption (one of the biggest concerns for datacenters) is even more telling.