Sizing Up Servers: Intel's Skylake-SP Xeon versus AMD's EPYC 7000 - The Server CPU Battle of the Decade?
by Johan De Gelas & Ian Cutress on July 11, 2017 12:15 PM EST- Posted in
- CPUs
- AMD
- Intel
- Xeon
- Enterprise
- Skylake
- Zen
- Naples
- Skylake-SP
- EPYC
Database Performance: MySQL Percona Server 5.7.0
For database benchmarking we still base our testing on Percona server 5.7, an enhanced drop-in replacement for MySQL. But we have updated our SQL benchmarking once again. This time we use Sysbench 1.0.7, which is a lot more efficient than the previous 0.4 and 0.5 versions. As a result, the measured numbers are quite a bit higher, especially on the strongest systems. So you cannot compare this with any similar Sysbench-based benchmarking we have done before.
For our testing we used the read-only OLTP benchmark, which is slightly less realistic, but still much more interesting than most other Sysbench tests. This allows us to measure CPU performance without creating an I/O bottleneck.
As expected, the EPYC 7601 can not deliver high database performance out of the box. A small database that can be mostly cached in the L3-cache is the worst case scenario for EPYC. That said, there are quite a few tuning opportunities on EPYC. According to AMD, if you enable Memory Interleaving, performance should rise a bit (+10-15%?). Unfortunately, a few days before our deadline our connection to the BMC failed, so we could not try it out. In a later article, we will go deeper into specific tuning for both platforms and test additional database systems.
Nevertheless, our point stands: out of the box is the EPYC CPU a rather mediocre transactional database CPU. With good tuning it is possible EPYC may pass the Xeon v4, but the 8176 is by far the champion here. It will be interesting to measure how EPYC compares in the non-transactional databases (Document stores, Key-value...) but transactional databases will remain Intel territory for now.
Typically when high response times were reported, this indicated low single threaded performance. However for EPYC this is not the case. We tested with a database that is quite a bit larger than the 8 MB L3-cache, and the high response time is probably a result of the L3-cache latency.
Facebook
Twitter
RSS
219 Comments
View All Comments
oldlaptop - Thursday, July 13, 2017 - link
Why on earth is gcc -Ofast being used to mimic "real-world", non-"aggressively optimized"(!) conditions? This is in fact the *most* aggressive optimization setting available; it is very sensitive to the exact program being compiled at best, and generates bloated (low priority on code size) and/or buggy code at worst (possibly even harming performance if the generated code is so big as to harm cache coherency). Most real-world software will be built with -O2 or possibly -Os. I can't help but wonder why questions weren't asked when SPEC complained about this unwisely aggressive optimization setting...peevee - Thursday, July 13, 2017 - link
"added a second full-blown 512 bit AVX-512 unit. "Do you mean "added second 256 ALU, which in combination with the first one implements full 512-bit AVX-512 unit"?
peevee - Thursday, July 13, 2017 - link
"getting data from the right top node to the bottom left node – should demand around 13 cycles. And before you get too concerned with that number, keep in mind that it compares very favorably with any off die communication that has to happen between different dies in (AMD's) Multi Chip Module (MCM), with the Skylake-SP's latency being around one-tenth of EPYC's."1/10th? Asking data from L3 on the chip next to it will take 130 (or even 65 if they are talking about averages) cycles? Does not sound realistic, you can request data from RAM at similar latencies already.
AmericasCup - Friday, July 14, 2017 - link
'For enterprises with a small infrastructure crew and server hardware on premise, spending time on hardware tuning is not an option most of the time.'Conversely, our small crew shop has been tuning AMD (selected for scalar floating point operations performance) for years. The experience and familiarity makes switching less attractive.
Also, you did all this in one week for AMD and two weeks for Intel? Did you ever sleep? KUDOS!
JohanAnandtech - Friday, July 21, 2017 - link
Thanks for appreciating the effort. Luckily, I got some help from Ian on Tuesday. :-)AntonErtl - Friday, July 14, 2017 - link
According to http://www.anandtech.com/show/10158/the-intel-xeon... if you execute just one AVX256 instruction on one core, this slows down the clocks of all E5v4 cores on the same socket for at least 1ms. Somewhere I read that newer Xeons only slow down the core that executes the AVX256 instruction. I expect that it works the same way for AVX512, and yes, this means that if you don't have a load with a heavy proportion of SIMD instructions, you are better off with AVX128 or SSE. The AMD variant of having only 128-bit FPUs and no clock slowdown looks better balanced to me. It might not win Linpack benchmark competitions, but for that one uses GPUs anyway these days.wagoo - Sunday, July 16, 2017 - link
Typo on the CLOSING THOUGHTS page: "dual Silver Xeon solutions" (dual socket)Great read though, thanks! Can finally replace my dual socket shanghai opteron home server soon :)
Chaser - Sunday, July 16, 2017 - link
AMD's CPU future is looking very promising!bongey - Tuesday, July 18, 2017 - link
EPYC power consumption is just wrong. Somehow you are 50W over what everyone else is getting at idle. https://www.servethehome.com/amd-epyc-7601-dual-so...Nenad - Thursday, July 20, 2017 - link
Interesting SPECint2006 results:- Intel in their slide #9 claims that Intel 8160 is 2% faster than EPYC 7601
- Anandtech in article tests that EPYC 7601 is 42% faster than Intel 8176
Those two are quite different, even if we ignore that 8176 should be faster than 8160. In other words, those Intel test results look very suspicious.