DDR5 Memory Cooling Performance For AMD EPYC Turin

AMD EPYC 9655P with 12 x DDR5-6000 memory comparing performance and thermals with Corsair Airflow memory cooling adapters. Benchmarks by Michael Larabel.

Compare your own system(s) to this result file with the Phoronix Test Suite by running the command: phoronix-test-suite benchmark 2411275-NE-MEMORYTHE74
Jump To Table - Results

View

Do Not Show Noisy Results
Do Not Show Results With Incomplete Data
Do Not Show Results With Little Change/Spread
List Notable Results
Show Result Confidence Charts
Allow Limiting Results To Certain Suite(s)

Statistics

Show Overall Harmonic Mean(s)
Show Overall Geometric Mean
Show Wins / Losses Counts (Pie Chart)
Normalize Results
Remove Outliers Before Calculating Averages

Graph Settings

Force Line Graphs Where Applicable
Convert To Scalar Where Applicable
Prefer Vertical Bar Graphs
No Box Plots
On Line Graphs With Missing Data, Connect The Line Gaps

Multi-Way Comparison

Condense Multi-Option Tests Into Single Result Graphs
Condense Test Profiles With Multiple Version Results Into Single Result Graphs

Table

Show Detailed System Result Table

Sensor Monitoring

Show Accumulated Sensor Monitoring Data For Displayed Results

Run Management

Highlight
Result
Toggle/Hide
Result
Result
Identifier
View Logs
Performance Per
Dollar
Date
Run
  Test
  Duration
Without Extra Cooling
November 27
  4 Hours, 38 Minutes
Corsair Airflow Memory Cooling
November 27
  3 Hours, 16 Minutes
Invert Behavior (Only Show Selected Data)
  3 Hours, 57 Minutes
Only show results matching title/arguments (delimit multiple options with a comma):
Do not show results matching title/arguments (delimit multiple options with a comma):


DDR5 Memory Cooling Performance For AMD EPYC TurinOpenBenchmarking.orgPhoronix Test SuiteAMD EPYC 9655P 96-Core @ 2.60GHz (96 Cores / 192 Threads)Supermicro Super Server H13SSL-N v1.01 (3.0 BIOS)AMD 1Ah12 x 64GB DDR5-6000MT/s Micron MTC40F2046S1RC64BDY QSFF3201GB Micron_7450_MTFDKCB3T2TFSASPEED2 x Broadcom NetXtreme BCM5720 PCIeUbuntu 24.106.12.0-rc7-linux-pm-next-phx (x86_64)GNOME Shell 47.0X ServerGCC 14.2.0ext41024x768ProcessorMotherboardChipsetMemoryDiskGraphicsNetworkOSKernelDesktopDisplay ServerCompilerFile-SystemScreen ResolutionDDR5 Memory Cooling Performance For AMD EPYC Turin BenchmarksSystem Logs- Transparent Huge Pages: madvise- --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --enable-bootstrap --enable-cet --enable-checking=release --enable-clocale=gnu --enable-default-pie --enable-gnu-unique-object --enable-languages=c,ada,c++,go,d,fortran,objc,obj-c++,m2,rust --enable-libphobos-checking=release --enable-libstdcxx-backtrace --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-link-serialization=2 --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-defaulted --enable-offload-targets=nvptx-none=/build/gcc-14-zdkDXv/gcc-14-14.2.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-14-zdkDXv/gcc-14-14.2.0/debian/tmp-gcn/usr --enable-plugin --enable-shared --enable-threads=posix --host=x86_64-linux-gnu --program-prefix=x86_64-linux-gnu- --target=x86_64-linux-gnu --with-abi=m64 --with-arch-32=i686 --with-build-config=bootstrap-lto-lean --with-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v - Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xb002116- OpenJDK Runtime Environment (build 21.0.5+11-Ubuntu-1ubuntu124.10)- Python 3.12.7- gather_data_sampling: Not affected + itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + reg_file_data_sampling: Not affected + retbleed: Not affected + spec_rstack_overflow: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced / Automatic IBRS; IBPB: conditional; STIBP: always-on; RSB filling; PBRSB-eIBRS: Not affected; BHI: Not affected + srbds: Not affected + tsx_async_abort: Not affected

Without Extra Cooling vs. Corsair Airflow Memory Cooling ComparisonPhoronix Test SuiteBaseline+24.5%+24.5%+49%+49%+73.5%+73.5%97.8%96.5%91.2%59.8%51.9%36.1%32.2%17.2%16.6%13.9%13.3%10.8%8.8%7.9%6.6%3.7%11.2%13.9%12.6%17.2%12.7%36%16.8%32.2%104 104 104 - 60144 144 144 - 60160 160 160 - 60d.M.M.S - Execution TimeS.w.1.0.6.AFalcon-7b-instruct-int4-ov - CPUGemma-7b-int4-ov - CPUPhi-3-mini-128k-instruct-int4-ov - CPUA.w.3.5.ATinyLlama-1.1B-Chat-v1.0 - CPUMPI CPU - water_GMX50_bareApache Spark Bayes2626d.M.M.S - Mesh TimeCompression Rating263%262.8%d.S.M.S - Execution Time2.4%TinyLlama-1.1B-Chat-v1.0 - CPU - T.T.F.TTinyLlama-1.1B-Chat-v1.0 - CPU - T.P.O.TPhi-3-mini-128k-instruct-int4-ov - CPU - T.T.F.TPhi-3-mini-128k-instruct-int4-ov - CPU - T.P.O.TFalcon-7b-instruct-int4-ov - CPU - T.T.F.TFalcon-7b-instruct-int4-ov - CPU - T.P.O.TGemma-7b-int4-ov - CPU - T.T.F.TGemma-7b-int4-ov - CPU - T.P.O.THigh Performance Conjugate GradientHigh Performance Conjugate GradientHigh Performance Conjugate GradientOpenFOAMNAMDOpenVINO GenAIOpenVINO GenAIOpenVINO GenAINAMDOpenVINO GenAIGROMACSRenaissanceGraph500Graph500OpenFOAM7-Zip CompressionGraph500Graph500OpenFOAMOpenVINO GenAIOpenVINO GenAIOpenVINO GenAIOpenVINO GenAIOpenVINO GenAIOpenVINO GenAIOpenVINO GenAIOpenVINO GenAIWithout Extra CoolingCorsair Airflow Memory Cooling

DDR5 Memory Cooling Performance For AMD EPYC Turinhpcg: 104 104 104 - 60hpcg: 144 144 144 - 60hpcg: 160 160 160 - 60graph500: 26graph500: 26graph500: 26graph500: 26openvino-genai: TinyLlama-1.1B-Chat-v1.0 - CPUopenvino-genai: TinyLlama-1.1B-Chat-v1.0 - CPU - Time To First Tokenopenvino-genai: TinyLlama-1.1B-Chat-v1.0 - CPU - Time Per Output Tokenopenvino-genai: Phi-3-mini-128k-instruct-int4-ov - CPUopenvino-genai: Phi-3-mini-128k-instruct-int4-ov - CPU - Time To First Tokenopenvino-genai: Phi-3-mini-128k-instruct-int4-ov - CPU - Time Per Output Tokenopenvino-genai: Falcon-7b-instruct-int4-ov - CPUopenvino-genai: Falcon-7b-instruct-int4-ov - CPU - Time To First Tokenopenvino-genai: Falcon-7b-instruct-int4-ov - CPU - Time Per Output Tokenopenvino-genai: Gemma-7b-int4-ov - CPUopenvino-genai: Gemma-7b-int4-ov - CPU - Time To First Tokenopenvino-genai: Gemma-7b-int4-ov - CPU - Time Per Output Tokennamd: ATPase with 327,506 Atomsnamd: STMV with 1,066,628 Atomsgromacs: MPI CPU - water_GMX50_bareopenfoam: drivaerFastback, Small Mesh Size - Mesh Timeopenfoam: drivaerFastback, Small Mesh Size - Execution Timeopenfoam: drivaerFastback, Medium Mesh Size - Mesh Timeopenfoam: drivaerFastback, Medium Mesh Size - Execution Timeblender: BMW27 - CPU-Onlyblender: Barbershop - CPU-Onlyblender: Junkshop - CPU-Onlycompress-7zip: Compression Ratingcompress-7zip: Decompression Ratingsvt-av1: Preset 13 - Bosphorus 4Ksvt-av1: Preset 13 - Beauty 4K 10-bitsvt-av1: Preset 3 - Bosphorus 4Ksvt-av1: Preset 3 - Beauty 4K 10-bitbuild-linux-kernel: defconfigbuild-linux-kernel: allmodconfigbuild2: Time To Compilerenaissance: Apache Spark Bayesnwchem: C240 BuckyballWithout Extra CoolingCorsair Airflow Memory Cooling31.669530.715430.67541446300000149134000049748400067122900058.0120.6917.2447.5227.1521.0537.5332.9126.6528.6742.1534.8810.657832.7815815.68623.74296723.316389110.45187313.8686613.18124.4916.86756280625450421.99717.14815.6901.84422.313195.49152.511205.51360.862.645560.356258.65251406690000144858000054102200072399300066.0718.6015.1455.6824.1117.9651.0629.2119.5937.9136.1026.3812.427044.2239617.77623.54769223.880443103.58124196.3824213.24124.2316.89784376626810425.02317.11115.5911.84222.385195.56653.202185.41354.7OpenBenchmarking.org

High Performance Conjugate Gradient

HPCG is the High Performance Conjugate Gradient and is a new scientific benchmark from Sandia National Lans focused for super-computer testing with modern real-world workloads compared to HPCC. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgGFLOP/s, More Is BetterHigh Performance Conjugate Gradient 3.1X Y Z: 104 104 104 - RT: 60Without Extra CoolingCorsair Airflow Memory Cooling1428425670SE +/- 0.57, N = 12SE +/- 0.10, N = 331.6762.651. (CXX) g++ options: -O3 -ffast-math -ftree-vectorize -lmpi_cxx -lmpi

OpenBenchmarking.orgGFLOP/s, More Is BetterHigh Performance Conjugate Gradient 3.1X Y Z: 144 144 144 - RT: 60Without Extra CoolingCorsair Airflow Memory Cooling1428425670SE +/- 0.00, N = 3SE +/- 0.74, N = 330.7260.361. (CXX) g++ options: -O3 -ffast-math -ftree-vectorize -lmpi_cxx -lmpi

OpenBenchmarking.orgGFLOP/s, More Is BetterHigh Performance Conjugate Gradient 3.1X Y Z: 160 160 160 - RT: 60Without Extra CoolingCorsair Airflow Memory Cooling1326395265SE +/- 0.01, N = 3SE +/- 0.20, N = 330.6858.651. (CXX) g++ options: -O3 -ffast-math -ftree-vectorize -lmpi_cxx -lmpi

Graph500

This is a benchmark of the reference implementation of Graph500, an HPC benchmark focused on data intensive loads and commonly tested on supercomputers for complex data problems. Graph500 primarily stresses the communication subsystem of the hardware under test. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgbfs median_TEPS, More Is BetterGraph500 3.0Scale: 26Without Extra CoolingCorsair Airflow Memory Cooling300M600M900M1200M1500M144630000014066900001. (CC) gcc options: -fcommon -O3 -lpthread -lm -lmpi

OpenBenchmarking.orgbfs max_TEPS, More Is BetterGraph500 3.0Scale: 26Without Extra CoolingCorsair Airflow Memory Cooling300M600M900M1200M1500M149134000014485800001. (CC) gcc options: -fcommon -O3 -lpthread -lm -lmpi

OpenBenchmarking.orgsssp median_TEPS, More Is BetterGraph500 3.0Scale: 26Without Extra CoolingCorsair Airflow Memory Cooling120M240M360M480M600M4974840005410220001. (CC) gcc options: -fcommon -O3 -lpthread -lm -lmpi

OpenBenchmarking.orgsssp max_TEPS, More Is BetterGraph500 3.0Scale: 26Without Extra CoolingCorsair Airflow Memory Cooling160M320M480M640M800M6712290007239930001. (CC) gcc options: -fcommon -O3 -lpthread -lm -lmpi

OpenVINO GenAI

OpenBenchmarking.orgtokens/s, More Is BetterOpenVINO GenAI 2024.5Model: TinyLlama-1.1B-Chat-v1.0 - Device: CPUWithout Extra CoolingCorsair Airflow Memory Cooling1530456075SE +/- 0.27, N = 3SE +/- 0.32, N = 358.0166.07

OpenBenchmarking.orgtokens/s, More Is BetterOpenVINO GenAI 2024.5Model: Phi-3-mini-128k-instruct-int4-ov - Device: CPUWithout Extra CoolingCorsair Airflow Memory Cooling1326395265SE +/- 0.35, N = 4SE +/- 0.09, N = 447.5255.68

OpenBenchmarking.orgtokens/s, More Is BetterOpenVINO GenAI 2024.5Model: Falcon-7b-instruct-int4-ov - Device: CPUWithout Extra CoolingCorsair Airflow Memory Cooling1224364860SE +/- 0.06, N = 3SE +/- 0.17, N = 337.5351.06

OpenBenchmarking.orgtokens/s, More Is BetterOpenVINO GenAI 2024.5Model: Gemma-7b-int4-ov - Device: CPUWithout Extra CoolingCorsair Airflow Memory Cooling918273645SE +/- 0.14, N = 3SE +/- 0.03, N = 328.6737.91

NAMD

OpenBenchmarking.orgns/day, More Is BetterNAMD 3.0Input: ATPase with 327,506 AtomsWithout Extra CoolingCorsair Airflow Memory Cooling3691215SE +/- 0.04, N = 7SE +/- 0.04, N = 710.6612.43

OpenBenchmarking.orgns/day, More Is BetterNAMD 3.0Input: STMV with 1,066,628 AtomsWithout Extra CoolingCorsair Airflow Memory Cooling0.95041.90082.85123.80164.752SE +/- 0.01539, N = 3SE +/- 0.00252, N = 42.781584.22396

GROMACS

The GROMACS (GROningen MAchine for Chemical Simulations) molecular dynamics package testing with the water_GMX50 data. This test profile allows selecting between CPU and GPU-based GROMACS builds. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgNs Per Day, More Is BetterGROMACS 2024Implementation: MPI CPU - Input: water_GMX50_bareWithout Extra CoolingCorsair Airflow Memory Cooling48121620SE +/- 0.02, N = 3SE +/- 0.02, N = 315.6917.781. (CXX) g++ options: -O3 -lm

OpenFOAM

OpenFOAM is the leading free, open-source software for computational fluid dynamics (CFD). This test profile currently uses the drivaerFastback test case for analyzing automotive aerodynamics or alternatively the older motorBike input. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenFOAM 10Input: drivaerFastback, Small Mesh Size - Mesh TimeWithout Extra CoolingCorsair Airflow Memory Cooling61218243023.7423.551. (CXX) g++ options: -std=c++14 -m64 -O3 -ftemplate-depth-100 -fPIC -fuse-ld=bfd -Xlinker --add-needed --no-as-needed -lfiniteVolume -lmeshTools -lparallel -llagrangian -lregionModels -lgenericPatchFields -lOpenFOAM -ldl -lm

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenFOAM 10Input: drivaerFastback, Small Mesh Size - Execution TimeWithout Extra CoolingCorsair Airflow Memory Cooling61218243023.3223.881. (CXX) g++ options: -std=c++14 -m64 -O3 -ftemplate-depth-100 -fPIC -fuse-ld=bfd -Xlinker --add-needed --no-as-needed -lfiniteVolume -lmeshTools -lparallel -llagrangian -lregionModels -lgenericPatchFields -lOpenFOAM -ldl -lm

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenFOAM 10Input: drivaerFastback, Medium Mesh Size - Mesh TimeWithout Extra CoolingCorsair Airflow Memory Cooling20406080100110.45103.581. (CXX) g++ options: -std=c++14 -m64 -O3 -ftemplate-depth-100 -fPIC -fuse-ld=bfd -Xlinker --add-needed --no-as-needed -lfiniteVolume -lmeshTools -lparallel -llagrangian -lregionModels -lgenericPatchFields -lOpenFOAM -ldl -lm

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenFOAM 10Input: drivaerFastback, Medium Mesh Size - Execution TimeWithout Extra CoolingCorsair Airflow Memory Cooling70140210280350313.87196.381. (CXX) g++ options: -std=c++14 -m64 -O3 -ftemplate-depth-100 -fPIC -fuse-ld=bfd -Xlinker --add-needed --no-as-needed -lfiniteVolume -lmeshTools -lparallel -llagrangian -lregionModels -lgenericPatchFields -lOpenFOAM -ldl -lm

Blender

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 4.3Blend File: BMW27 - Compute: CPU-OnlyWithout Extra CoolingCorsair Airflow Memory Cooling3691215SE +/- 0.05, N = 4SE +/- 0.04, N = 413.1813.24

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 4.3Blend File: Barbershop - Compute: CPU-OnlyWithout Extra CoolingCorsair Airflow Memory Cooling306090120150SE +/- 0.24, N = 3SE +/- 0.20, N = 3124.49124.23

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 4.3Blend File: Junkshop - Compute: CPU-OnlyWithout Extra CoolingCorsair Airflow Memory Cooling48121620SE +/- 0.06, N = 3SE +/- 0.15, N = 316.8616.89

7-Zip Compression

OpenBenchmarking.orgMIPS, More Is Better7-Zip Compression 24.05Test: Compression RatingWithout Extra CoolingCorsair Airflow Memory Cooling200K400K600K800K1000KSE +/- 4691.61, N = 3SE +/- 4235.99, N = 37562807843761. (CXX) g++ options: -lpthread -ldl -O2 -fPIC

OpenBenchmarking.orgMIPS, More Is Better7-Zip Compression 24.05Test: Decompression RatingWithout Extra CoolingCorsair Airflow Memory Cooling130K260K390K520K650KSE +/- 766.64, N = 3SE +/- 486.33, N = 36254506268101. (CXX) g++ options: -lpthread -ldl -O2 -fPIC

SVT-AV1

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 2.3Encoder Mode: Preset 13 - Input: Bosphorus 4KWithout Extra CoolingCorsair Airflow Memory Cooling90180270360450SE +/- 10.38, N = 15SE +/- 10.67, N = 15422.00425.021. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 2.3Encoder Mode: Preset 13 - Input: Beauty 4K 10-bitWithout Extra CoolingCorsair Airflow Memory Cooling48121620SE +/- 0.00, N = 3SE +/- 0.01, N = 317.1517.111. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 2.3Encoder Mode: Preset 3 - Input: Bosphorus 4KWithout Extra CoolingCorsair Airflow Memory Cooling48121620SE +/- 0.06, N = 3SE +/- 0.07, N = 315.6915.591. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 2.3Encoder Mode: Preset 3 - Input: Beauty 4K 10-bitWithout Extra CoolingCorsair Airflow Memory Cooling0.41490.82981.24471.65962.0745SE +/- 0.008, N = 3SE +/- 0.004, N = 31.8441.8421. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq

Timed Linux Kernel Compilation

This test times how long it takes to build the Linux kernel in a default configuration (defconfig) for the architecture being tested or alternatively an allmodconfig for building all possible kernel modules for the build. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Linux Kernel Compilation 6.8Build: defconfigWithout Extra CoolingCorsair Airflow Memory Cooling510152025SE +/- 0.23, N = 3SE +/- 0.22, N = 322.3122.39

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Linux Kernel Compilation 6.8Build: allmodconfigWithout Extra CoolingCorsair Airflow Memory Cooling4080120160200SE +/- 0.38, N = 3SE +/- 0.35, N = 3195.49195.57

Build2

OpenBenchmarking.orgSeconds, Fewer Is BetterBuild2 0.17Time To CompileWithout Extra CoolingCorsair Airflow Memory Cooling1224364860SE +/- 0.39, N = 3SE +/- 0.11, N = 352.5153.20

Renaissance

OpenBenchmarking.orgms, Fewer Is BetterRenaissance 0.16Test: Apache Spark BayesWithout Extra CoolingCorsair Airflow Memory Cooling50100150200250SE +/- 1.44, N = 3SE +/- 0.13, N = 3205.5185.4MIN: 185.67 / MAX: 247.59MIN: 160 / MAX: 259.09

NWChem

OpenBenchmarking.orgSeconds, Fewer Is BetterNWChem 7.2.3Input: C240 BuckyballWithout Extra CoolingCorsair Airflow Memory Cooling300600900120015001360.81354.71. (F9X) gfortran options: -lnwctask -lccsd -lmcscf -lselci -lmp2 -lmoints -lstepper -ldriver -loptim -lnwdft -lgradients -lcphf -lesp -lddscf -ldangchang -lguess -lhessian -lvib -lnwcutil -lrimp2 -lproperty -lsolvation -lnwints -lprepar -lnwmd -lnwpw -lofpw -lpaw -lpspw -lband -lnwpwlib -lcafe -lspace -lanalyze -lqhop -lpfft -ldplot -ldrdy -lvscf -lqmmm -lqmd -letrans -ltce -lbq -lmm -lcons -lperfm -ldntmc -lccca -ldimqm -lfcidump -lgwmol -lga -larmci -lpeigs -l64to32 -llapack -lopenblas -lpthread -lrt -lcomex -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz -ffast-math -std=legacy -fdefault-integer-8 -O0