POWER9 Benchmarks vs. Intel Xeon vs. AMD EPYC Performance On Debian Linux

[phoronix]

Phoronix: POWER9 Benchmarks vs. Intel Xeon vs. AMD EPYC Performance On Debian Linux

For several days we've had remote access to one of the brand new Raptor Talos II Workstations that is powered by POWER9 processors and open-source down through the firmware. For those curious how these latest POWER processors compare to AMD EPYC and Intel Xeon processors, here are some benchmarks comparing against of the few other systems in house while all testing was done from Debian GNU/Linux.

http://www.phoronix.com/vr.php?view=26179

[soulsource]

A possible explanation for the comparatively low performance with multimedia encoders is that they might use SIMD extensions for x86 and fall back portable C code on other platforms.
I haven't checked their source code, so this is just a kinda educated guess though.

[sireangelus]

so... completely overpriced useless hardware?

[mastermind]

Not surprising, given that probably millions of man-hours went into x86 compilers' backends and specific code optimization over decades. (multimedia encoders are likely hand-optimized indeed) I for one can't wait to get my hands on one of these Power9 systems.

[edwaleni]

I checked Talos site. Just under $10k for this configuration.

[chuckula]

I checked Talos site. Just under $10k for this configuration.

And people scream bloody murder about spending $2500 for a nicely equipped Core i9 system or $3500 if you want to go all the way up to the 7980XE.

[cybertraveler]

Really great benchmarks Michael, but I think including some price/cost information with the systems is essential. Without knowing what the cost differences are between the systems, the data isn't nearly as useful. If it turned out that the Intel system cost almost the same as the AMD and Talos systems, then its comparatively good performance makes it stand out as a very good option for a consumer. If however the Intel system cost four times as much as the AMD or Talos, then its performance would be considered bad, as the Intel system wasn't producing four times the performance.

Even basic info at the top of the article would be useful, eg:
CPU cost
Mobo cost
RAM cost

... or just the costs of the complete systems.

It doesn't really matter what currency the costs are listed in.

[edwaleni]

And people scream bloody murder about spending $2500 for a nicely equipped Core i9 system or $3500 if you want to go all the way up to the 7980XE.

If Talos could produce several 100's of thousands of these, I am sure they can get the price down considerably. Until then, this is a near custom fab.

Starship probably has more current info, but I think most planar fabs have a 10,000 unit minimum for an order like this.

[Michael]

Really great benchmarks Michael, but I think including some price/cost information with the systems is essential. Without knowing what the cost differences are between the systems, the data isn't nearly as useful. If it turned out that the Intel system cost almost the same as the AMD and Talos systems, then its comparatively good performance makes it stand out as a very good option for a consumer. If however the Intel system cost four times as much as the AMD or Talos, then its performance would be considered bad, as the Intel system wasn't producing four times the performance.

Even basic info at the top of the article would be useful, eg:
CPU cost
Mobo cost
RAM cost

... or just the costs of the complete systems.

It doesn't really matter what currency the costs are listed in.

While mentioned why I didn't include price/Watt, I opted to not include price/dollar because it's really tough to get an accurate comparison of the total cost of ownership.... Since these systems tested didn't have identical storage, the Tyan servers had redundant power supplies, different amounts of RAM, etc. So would have been messy trying to compare the total cost of ownership in a fully comparable manner. Especially in the server/workstation market where many customers get discounts from their sales when ordering in any volumes, etc. If just using the CPU price itself, that isn't entirely reflective either with there being many more EPYC/Xeon motherboards available at lower cost than POWER9, so really looking then at just one aspect.

[WolfpackN64]

If they would produce a more affordable single socket system, I might consider it. For now, it's just too expensive.