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.

Raptor Computing System's Talos II Secure Workstation is backed by up to two IBM POWER9 processors, supports five PCI Express slots with PCI-E 4.0 support, and is fully open down through the firmware of the system. The Talos II is also a made in the US system with the CPUs being manufactured in New York and the board and systems being manufactured in Texas.

Pictures courtesy Raptor Computing Systems.

The system we were provided remote access with is a dual eight-core Talos II system, with four threads per core yielding a combined total of 64 threads. Raptor is offering up to 22-core CPUs (88 threads per CPU) and they say in the coming weeks we should be able to get remote access for testing those higher-end processors as well with dual 22-cores (176 threads). Raptor Engineering is also working on overclocking tools as well for being able to overclock these POWER9 systems.

I don't have any crypto-currency mining benchmarks for this article but figures provided by Raptor indicate around 2.7kH/s for Monero mining or 2.8kH/s for Cryptonight while pulling 405 Watts while closer to 325 Watts without SAS and other peripherals.

Under full load the system in its configuration pulls roughly 430 Watts AC power while under idle is around 110 Watts. I don't have any performance-per-Watt benchmarks though due to only remote access and not having any WattsUp Pro or similar for being able to interface with PTS for the performance-per-Watt benchmarks.

The Raptor Talos II test system was configured with the aforementioned dual eight core IBM POWER9 CPUs yielding 64 threads, 256GB of system memory, a 500GB HDD (what they had available, but I/O testing thus isn't a focus for today's tests), and an AMD Radeon Pro WX 7100 graphics card. The POWER9 CPUs clock up to 3.80GHz.

The system was loaded with Debian Testing and using the Linux 4.16 PPC64LE kernel. The Linux 4.15 kernel and newer currently support the Talos II. Our Xeon/EPYC testing was done with Debian Testing as well using its GCC 7.3 stock compiler, EXT4 file-system, and upgrading to the Linux 4.16 kernel.

The other x86 server/workstation systems I had locally for comparison to these initial POWER9 benchmark numbers were:

2 x Intel Xeon Gold 6138 - The Tyan S7106 1U barebones with two Xeon Gold 6138 processors yielding a combined 40 cores / 80 threads, 96GB of RAM, and Samsung 850 EVO 256GB. Xeon Gold 6138 processors have a 2.0GHz base frequency and 3.7GHz turbo frequency.

AMD EPYC 7551 - The Gigabyte MZ31-AR0 with an AMD EPYC 7551, which is 32 cores / 64 threads at a 2.0GHz base clock frequency, 2.55GHz all core boost speed, and 3.0GHz maximum boost clock speed. This system had 32GB of memory and NVIDIA GeForce GTX 970 graphics.

AMD EPYC 7601 - The TYAN B8026T70AE24HR with an AMD EPYC 7601, which is 32 cores / 64 threads with a 2.2GHz base frequency, 2.7GHz all core boost frequency, and 3.2GHz maximum boost clock speed. This system had 128GB of DDR4 and a 280GB Intel 900p Optane SSD.

Those were the systems available for this initial round of testing against the 64-thread Talos II Workstation while all systems were on Debian Testing with the Linux 4.16 kernel. This is just the raw performance results due to not being able to deliver performance-per-Watt metrics at this time from the Talos II.