AMD INVLPGB Linux Patch Performance

Benchmarks for a future article looking at AMD broadcast TLB invalidation Linux kernel patches with the INVLPGB instruction on newer AMD Zen 3 processors.

HTML result view exported from: https://openbenchmarking.org/result/2412250-NE-AMDINVLPG56&grs&rdt.

AMD INVLPGB Linux Patch PerformanceProcessorMotherboardChipsetMemoryDiskGraphicsNetworkOSKernelDesktopDisplay ServerCompilerFile-SystemScreen ResolutionINVLPGB PatchedLinux 6.13 GitAMD 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_MTFDKCB3T2TFS + 257GB Flash DriveASPEED2 x Broadcom NetXtreme BCM5720 PCIeUbuntu 24.106.13.0-rc4-phx-broadcast-tlb (x86_64)GNOME Shell 47.0X ServerGCC 14.2.0ext41024x7686.13.0-rc4-phx-stock (x86_64)OpenBenchmarking.orgKernel Details- Transparent Huge Pages: madviseCompiler Details- --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 Disk Details- NONE / relatime,rw,stripe=64 / Block Size: 4096Processor Details- Scaling Governor: acpi-cpufreq performance (Boost: Enabled) - CPU Microcode: 0xb002116Java Details- OpenJDK Runtime Environment (build 21.0.5+11-Ubuntu-1ubuntu124.10)Python Details- Python 3.12.7Security Details- 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

AMD INVLPGB Linux Patch Performancedacapobench: Zxing 1D/2D Barcode Image Processingdacapobench: Jythonmariadb: oltp_update_index - 256openfoam: drivaerFastback, Medium Mesh Size - Mesh Timecassandra: Writesapache-iotdb: 800 - 100 - 800 - 400nginx: 500quicksilver: CORAL2 P1whisper-cpp: ggml-small.en - 2016 State of the Unionnpb: EP.Capache-iotdb: 800 - 100 - 800 - 400clickhouse: 100M Rows Hits Dataset, Second Runapache-iotdb: 500 - 100 - 500 - 100dacapobench: GraphChispeedb: Read Rand Write Randwhisper-cpp: ggml-medium.en - 2016 State of the Unionrenaissance: Savina Reactors.IOapache-iotdb: 500 - 100 - 800 - 400rodinia: OpenMP Leukocytedacapobench: Avrora AVR Simulation Frameworkclickhouse: 100M Rows Hits Dataset, Third Runapache-iotdb: 500 - 100 - 500 - 100renaissance: Apache Spark PageRankapache-iotdb: 500 - 100 - 800 - 100apache-iotdb: 800 - 100 - 500 - 100renaissance: Apache Spark Bayesnpb: BT.Capache-iotdb: 800 - 100 - 500 - 100renaissance: ALS Movie Lensspeedb: Rand Readapache-iotdb: 500 - 100 - 800 - 100apache-iotdb: 500 - 100 - 500 - 400npb: SP.Cmariadb: oltp_read_write - 256rocksdb: Read While Writingbuild-python: Defaultapache-iotdb: 500 - 100 - 800 - 400dacapobench: BioJava Biological Data Frameworkbuild-python: Released Build, PGO + LTO Optimizedclickhouse: 100M Rows Hits Dataset, First Run / Cold Cachedacapobench: Apache Xalan XSLTmariadb: oltp_read_only - 256speedb: Update Randapache-iotdb: 800 - 100 - 500 - 400build-llvm: Ninjamemcached: 1:5blender: Barbershop - CPU-Onlynpb: LU.Cbuild-llvm: Unix Makefilesapache-iotdb: 500 - 100 - 500 - 400npb: IS.Dbuild-godot: Time To Compilenamd: STMV with 1,066,628 Atomsdacapobench: Apache Kafkadacapobench: jMonkeyEnginerenaissance: In-Memory Database Shootoutnamd: ATPase with 327,506 Atomsquicksilver: CORAL2 P2xmrig: GhostRider - 1Mrelion: Basic - CPUnpb: MG.CINVLPGB PatchedLinux 6.13 Git6194158117679106.20884452025206.65517615.5434996667219.8737511450.17144092202749.6247.0323233106319453.069965651.4214.4831.2982556759.44977513772283.360.9340.04181.7356244.9111889017018368.8627286939121681953100250585155286.471737651473755012.9841246313544547191.972735.328123786852223812072601090.7753583145.46124.48311722.61161.060167.956984.6580.8314.20973505068064512.912.221252205666716061.4211.380147724.666484320113658109.86392438419212.98503771.7634086667225.4195211180.11141171394764.7147.8523633054826460.621295722.9211.8131.6742585751.84968068092261.361.4940.40180.1353192.1611788721018214.862208408312073844999520170154232.791726151464188913.0671254230554519190.912738.998163768851986712123788490.4153569817.66124.03312554.83160.666168.367001.4280.6874.21510504668084511.612.224432205666713706.9214.442145177.79OpenBenchmarking.org

DaCapo Benchmark

Java Test: Zxing 1D/2D Barcode Image Processing

OpenBenchmarking.orgmsec, Fewer Is BetterDaCapo Benchmark 23.11Java Test: Zxing 1D/2D Barcode Image ProcessingINVLPGB PatchedLinux 6.13 Git140280420560700SE +/- 6.43, N = 3SE +/- 4.04, N = 3619648

DaCapo Benchmark

Java Test: Jython

OpenBenchmarking.orgmsec, Fewer Is BetterDaCapo Benchmark 23.11Java Test: JythonINVLPGB PatchedLinux 6.13 Git9001800270036004500SE +/- 10.53, N = 3SE +/- 37.01, N = 1541584320

MariaDB

Test: oltp_update_index - Threads: 256

OpenBenchmarking.orgQueries Per Second, More Is BetterMariaDB 11.5Test: oltp_update_index - Threads: 256INVLPGB PatchedLinux 6.13 Git30K60K90K120K150KSE +/- 200.13, N = 3SE +/- 190.75, N = 31176791136581. (CXX) g++ options: -pie -fPIC -fstack-protector -O3 -lnuma -lpcre2-8 -lcrypt -laio -lz -lm -lssl -lcrypto -lpthread -ldl

OpenFOAM

Input: drivaerFastback, Medium Mesh Size - Mesh Time

OpenBenchmarking.orgSeconds, Fewer Is BetterOpenFOAM 10Input: drivaerFastback, Medium Mesh Size - Mesh TimeINVLPGB PatchedLinux 6.13 Git20406080100106.21109.861. (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

Apache Cassandra

Test: Writes

OpenBenchmarking.orgOp/s, More Is BetterApache Cassandra 5.0Test: WritesINVLPGB PatchedLinux 6.13 Git100K200K300K400K500KSE +/- 2371.91, N = 3SE +/- 5248.18, N = 3452025438419

Apache IoTDB

Device Count: 800 - Batch Size Per Write: 100 - Sensor Count: 800 - Client Number: 400

OpenBenchmarking.orgAverage Latency, Fewer Is BetterApache IoTDB 1.2Device Count: 800 - Batch Size Per Write: 100 - Sensor Count: 800 - Client Number: 400INVLPGB PatchedLinux 6.13 Git50100150200250SE +/- 3.85, N = 3SE +/- 2.18, N = 3206.65212.98MAX: 26883.6MAX: 26561.37

nginx

Connections: 500

OpenBenchmarking.orgRequests Per Second, More Is Betternginx 1.23.2Connections: 500INVLPGB PatchedLinux 6.13 Git110K220K330K440K550KSE +/- 1146.82, N = 3SE +/- 1734.77, N = 3517615.54503771.761. (CC) gcc options: -lluajit-5.1 -lm -lssl -lcrypto -lpthread -ldl -std=c99 -O2

Quicksilver

Input: CORAL2 P1

OpenBenchmarking.orgFigure Of Merit, More Is BetterQuicksilver 20230818Input: CORAL2 P1INVLPGB PatchedLinux 6.13 Git7M14M21M28M35MSE +/- 98713.95, N = 3SE +/- 46666.67, N = 334996667340866671. (CXX) g++ options: -fopenmp -O3 -march=native

Whisper.cpp

Model: ggml-small.en - Input: 2016 State of the Union

OpenBenchmarking.orgSeconds, Fewer Is BetterWhisper.cpp 1.6.2Model: ggml-small.en - Input: 2016 State of the UnionINVLPGB PatchedLinux 6.13 Git50100150200250SE +/- 1.60, N = 3SE +/- 0.39, N = 3219.87225.421. (CXX) g++ options: -O3 -std=c++11 -fPIC -pthread -msse3 -mssse3 -mavx -mf16c -mfma -mavx2 -mavx512f -mavx512cd -mavx512vl -mavx512dq -mavx512bw -mavx512vbmi -mavx512vnni

NAS Parallel Benchmarks

Test / Class: EP.C

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: EP.CINVLPGB PatchedLinux 6.13 Git2K4K6K8K10KSE +/- 98.82, N = 15SE +/- 27.23, N = 311450.1711180.111. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz 2. Open MPI 4.1.6

Apache IoTDB

Device Count: 800 - Batch Size Per Write: 100 - Sensor Count: 800 - Client Number: 400

OpenBenchmarking.orgpoint/sec, More Is BetterApache IoTDB 1.2Device Count: 800 - Batch Size Per Write: 100 - Sensor Count: 800 - Client Number: 400INVLPGB PatchedLinux 6.13 Git30M60M90M120M150MSE +/- 1488492.23, N = 3SE +/- 870384.94, N = 3144092202141171394

ClickHouse

100M Rows Hits Dataset, Second Run

OpenBenchmarking.orgQueries Per Minute, Geo Mean, More Is BetterClickHouse 22.12.3.5100M Rows Hits Dataset, Second RunINVLPGB PatchedLinux 6.13 Git160320480640800SE +/- 1.96, N = 3SE +/- 4.01, N = 3749.62764.71MIN: 69.28 / MAX: 6666.67MIN: 69.85 / MAX: 8571.43

Apache IoTDB

Device Count: 500 - Batch Size Per Write: 100 - Sensor Count: 500 - Client Number: 100

OpenBenchmarking.orgAverage Latency, Fewer Is BetterApache IoTDB 1.2Device Count: 500 - Batch Size Per Write: 100 - Sensor Count: 500 - Client Number: 100INVLPGB PatchedLinux 6.13 Git1122334455SE +/- 0.36, N = 3SE +/- 0.55, N = 347.0347.85MAX: 12578.34MAX: 12576.32

DaCapo Benchmark

Java Test: GraphChi

OpenBenchmarking.orgmsec, Fewer Is BetterDaCapo Benchmark 23.11Java Test: GraphChiINVLPGB PatchedLinux 6.13 Git5001000150020002500SE +/- 15.84, N = 3SE +/- 8.96, N = 323232363

Speedb

Test: Read Random Write Random

OpenBenchmarking.orgOp/s, More Is BetterSpeedb 2.7Test: Read Random Write RandomINVLPGB PatchedLinux 6.13 Git700K1400K2100K2800K3500KSE +/- 22974.88, N = 3SE +/- 5973.22, N = 3310631930548261. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti

Whisper.cpp

Model: ggml-medium.en - Input: 2016 State of the Union

OpenBenchmarking.orgSeconds, Fewer Is BetterWhisper.cpp 1.6.2Model: ggml-medium.en - Input: 2016 State of the UnionINVLPGB PatchedLinux 6.13 Git100200300400500SE +/- 0.73, N = 3SE +/- 3.15, N = 3453.07460.621. (CXX) g++ options: -O3 -std=c++11 -fPIC -pthread -msse3 -mssse3 -mavx -mf16c -mfma -mavx2 -mavx512f -mavx512cd -mavx512vl -mavx512dq -mavx512bw -mavx512vbmi -mavx512vnni

Renaissance

Test: Savina Reactors.IO

OpenBenchmarking.orgms, Fewer Is BetterRenaissance 0.16Test: Savina Reactors.IOINVLPGB PatchedLinux 6.13 Git12002400360048006000SE +/- 52.24, N = 7SE +/- 75.13, N = 35651.45722.9MIN: 5405.73 / MAX: 9193.8MIN: 5596.11 / MAX: 9196.98

Apache IoTDB

Device Count: 500 - Batch Size Per Write: 100 - Sensor Count: 800 - Client Number: 400

OpenBenchmarking.orgAverage Latency, Fewer Is BetterApache IoTDB 1.2Device Count: 500 - Batch Size Per Write: 100 - Sensor Count: 800 - Client Number: 400INVLPGB PatchedLinux 6.13 Git50100150200250SE +/- 0.65, N = 3SE +/- 2.05, N = 3214.48211.81MAX: 26489.9MAX: 26536.58

Rodinia

Test: OpenMP Leukocyte

OpenBenchmarking.orgSeconds, Fewer Is BetterRodinia 3.1Test: OpenMP LeukocyteINVLPGB PatchedLinux 6.13 Git714212835SE +/- 0.16, N = 3SE +/- 0.33, N = 331.3031.671. (CXX) g++ options: -O2 -lOpenCL

DaCapo Benchmark

Java Test: Avrora AVR Simulation Framework

OpenBenchmarking.orgmsec, Fewer Is BetterDaCapo Benchmark 23.11Java Test: Avrora AVR Simulation FrameworkINVLPGB PatchedLinux 6.13 Git6001200180024003000SE +/- 19.37, N = 3SE +/- 8.50, N = 325562585

ClickHouse

100M Rows Hits Dataset, Third Run

OpenBenchmarking.orgQueries Per Minute, Geo Mean, More Is BetterClickHouse 22.12.3.5100M Rows Hits Dataset, Third RunINVLPGB PatchedLinux 6.13 Git160320480640800SE +/- 3.07, N = 3SE +/- 0.95, N = 3759.44751.84MIN: 69.04 / MAX: 8571.43MIN: 70.34 / MAX: 6666.67

Apache IoTDB

Device Count: 500 - Batch Size Per Write: 100 - Sensor Count: 500 - Client Number: 100

OpenBenchmarking.orgpoint/sec, More Is BetterApache IoTDB 1.2Device Count: 500 - Batch Size Per Write: 100 - Sensor Count: 500 - Client Number: 100INVLPGB PatchedLinux 6.13 Git20M40M60M80M100MSE +/- 291260.63, N = 3SE +/- 1081476.60, N = 39775137796806809

Renaissance

Test: Apache Spark PageRank

OpenBenchmarking.orgms, Fewer Is BetterRenaissance 0.16Test: Apache Spark PageRankINVLPGB PatchedLinux 6.13 Git5001000150020002500SE +/- 22.38, N = 3SE +/- 9.21, N = 32283.32261.3MIN: 1561.41 / MAX: 2328.07MIN: 1572.27 / MAX: 2279.05

Apache IoTDB

Device Count: 500 - Batch Size Per Write: 100 - Sensor Count: 800 - Client Number: 100

OpenBenchmarking.orgAverage Latency, Fewer Is BetterApache IoTDB 1.2Device Count: 500 - Batch Size Per Write: 100 - Sensor Count: 800 - Client Number: 100INVLPGB PatchedLinux 6.13 Git1428425670SE +/- 0.25, N = 3SE +/- 0.36, N = 360.9361.49MAX: 11306.84MAX: 11313.55

Apache IoTDB

Device Count: 800 - Batch Size Per Write: 100 - Sensor Count: 500 - Client Number: 100

OpenBenchmarking.orgAverage Latency, Fewer Is BetterApache IoTDB 1.2Device Count: 800 - Batch Size Per Write: 100 - Sensor Count: 500 - Client Number: 100INVLPGB PatchedLinux 6.13 Git918273645SE +/- 0.04, N = 3SE +/- 0.37, N = 340.0440.40MAX: 23821.12MAX: 23837.32

Renaissance

Test: Apache Spark Bayes

OpenBenchmarking.orgms, Fewer Is BetterRenaissance 0.16Test: Apache Spark BayesINVLPGB PatchedLinux 6.13 Git4080120160200SE +/- 1.59, N = 3SE +/- 0.85, N = 3181.7180.1MIN: 164.62 / MAX: 284.35MIN: 160.23 / MAX: 297.68

NAS Parallel Benchmarks

Test / Class: BT.C

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: BT.CINVLPGB PatchedLinux 6.13 Git80K160K240K320K400KSE +/- 3962.64, N = 5SE +/- 3855.67, N = 3356244.91353192.161. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz 2. Open MPI 4.1.6

Apache IoTDB

Device Count: 800 - Batch Size Per Write: 100 - Sensor Count: 500 - Client Number: 100

OpenBenchmarking.orgpoint/sec, More Is BetterApache IoTDB 1.2Device Count: 800 - Batch Size Per Write: 100 - Sensor Count: 500 - Client Number: 100INVLPGB PatchedLinux 6.13 Git30M60M90M120M150MSE +/- 226557.07, N = 3SE +/- 1087988.19, N = 3118890170117887210

Renaissance

Test: ALS Movie Lens

OpenBenchmarking.orgms, Fewer Is BetterRenaissance 0.16Test: ALS Movie LensINVLPGB PatchedLinux 6.13 Git4K8K12K16K20KSE +/- 114.89, N = 3SE +/- 45.86, N = 318368.818214.8MIN: 17674.24 / MAX: 18580.57MIN: 17691.46 / MAX: 18303.77

Speedb

Test: Random Read

OpenBenchmarking.orgOp/s, More Is BetterSpeedb 2.7Test: Random ReadINVLPGB PatchedLinux 6.13 Git130M260M390M520M650MSE +/- 1540848.02, N = 3SE +/- 7230810.31, N = 36272869396220840831. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti

Apache IoTDB

Device Count: 500 - Batch Size Per Write: 100 - Sensor Count: 800 - Client Number: 100

OpenBenchmarking.orgpoint/sec, More Is BetterApache IoTDB 1.2Device Count: 500 - Batch Size Per Write: 100 - Sensor Count: 800 - Client Number: 100INVLPGB PatchedLinux 6.13 Git30M60M90M120M150MSE +/- 67599.70, N = 3SE +/- 58310.64, N = 3121681953120738449

Apache IoTDB

Device Count: 500 - Batch Size Per Write: 100 - Sensor Count: 500 - Client Number: 400

OpenBenchmarking.orgpoint/sec, More Is BetterApache IoTDB 1.2Device Count: 500 - Batch Size Per Write: 100 - Sensor Count: 500 - Client Number: 400INVLPGB PatchedLinux 6.13 Git20M40M60M80M100MSE +/- 97805.09, N = 3SE +/- 603733.82, N = 310025058599520170

NAS Parallel Benchmarks

Test / Class: SP.C

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: SP.CINVLPGB PatchedLinux 6.13 Git30K60K90K120K150KSE +/- 279.05, N = 3SE +/- 1469.84, N = 3155286.47154232.791. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz 2. Open MPI 4.1.6

MariaDB

Test: oltp_read_write - Threads: 256

OpenBenchmarking.orgQueries Per Second, More Is BetterMariaDB 11.5Test: oltp_read_write - Threads: 256INVLPGB PatchedLinux 6.13 Git40K80K120K160K200KSE +/- 93.11, N = 3SE +/- 73.62, N = 31737651726151. (CXX) g++ options: -pie -fPIC -fstack-protector -O3 -lnuma -lpcre2-8 -lcrypt -laio -lz -lm -lssl -lcrypto -lpthread -ldl

RocksDB

Test: Read While Writing

OpenBenchmarking.orgOp/s, More Is BetterRocksDB 9.0Test: Read While WritingINVLPGB PatchedLinux 6.13 Git3M6M9M12M15MSE +/- 155717.24, N = 15SE +/- 157207.18, N = 314737550146418891. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti

Timed CPython Compilation

Build Configuration: Default

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed CPython Compilation 3.10.6Build Configuration: DefaultINVLPGB PatchedLinux 6.13 Git369121512.9813.07

Apache IoTDB

Device Count: 500 - Batch Size Per Write: 100 - Sensor Count: 800 - Client Number: 400

OpenBenchmarking.orgpoint/sec, More Is BetterApache IoTDB 1.2Device Count: 500 - Batch Size Per Write: 100 - Sensor Count: 800 - Client Number: 400INVLPGB PatchedLinux 6.13 Git30M60M90M120M150MSE +/- 332877.17, N = 3SE +/- 352556.17, N = 3124631354125423055

DaCapo Benchmark

Java Test: BioJava Biological Data Framework

OpenBenchmarking.orgmsec, Fewer Is BetterDaCapo Benchmark 23.11Java Test: BioJava Biological Data FrameworkINVLPGB PatchedLinux 6.13 Git10002000300040005000SE +/- 17.52, N = 3SE +/- 20.17, N = 345474519

Timed CPython Compilation

Build Configuration: Released Build, PGO + LTO Optimized

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed CPython Compilation 3.10.6Build Configuration: Released Build, PGO + LTO OptimizedINVLPGB PatchedLinux 6.13 Git4080120160200191.97190.91

ClickHouse

100M Rows Hits Dataset, First Run / Cold Cache

OpenBenchmarking.orgQueries Per Minute, Geo Mean, More Is BetterClickHouse 22.12.3.5100M Rows Hits Dataset, First Run / Cold CacheINVLPGB PatchedLinux 6.13 Git160320480640800SE +/- 4.99, N = 3SE +/- 6.88, N = 3735.32738.99MIN: 68.03 / MAX: 7500MIN: 69.85 / MAX: 7500

DaCapo Benchmark

Java Test: Apache Xalan XSLT

OpenBenchmarking.orgmsec, Fewer Is BetterDaCapo Benchmark 23.11Java Test: Apache Xalan XSLTINVLPGB PatchedLinux 6.13 Git2004006008001000SE +/- 2.00, N = 3SE +/- 7.60, N = 6812816

MariaDB

Test: oltp_read_only - Threads: 256

OpenBenchmarking.orgQueries Per Second, More Is BetterMariaDB 11.5Test: oltp_read_only - Threads: 256INVLPGB PatchedLinux 6.13 Git8K16K24K32K40KSE +/- 112.63, N = 3SE +/- 17.06, N = 337868376881. (CXX) g++ options: -pie -fPIC -fstack-protector -O3 -lnuma -lpcre2-8 -lcrypt -laio -lz -lm -lssl -lcrypto -lpthread -ldl

Speedb

Test: Update Random

OpenBenchmarking.orgOp/s, More Is BetterSpeedb 2.7Test: Update RandomINVLPGB PatchedLinux 6.13 Git110K220K330K440K550KSE +/- 726.96, N = 3SE +/- 210.08, N = 35222385198671. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti

Apache IoTDB

Device Count: 800 - Batch Size Per Write: 100 - Sensor Count: 500 - Client Number: 400

OpenBenchmarking.orgpoint/sec, More Is BetterApache IoTDB 1.2Device Count: 800 - Batch Size Per Write: 100 - Sensor Count: 500 - Client Number: 400INVLPGB PatchedLinux 6.13 Git30M60M90M120M150MSE +/- 1093112.93, N = 3SE +/- 875042.28, N = 3120726010121237884

Timed LLVM Compilation

Build System: Ninja

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed LLVM Compilation 16.0Build System: NinjaINVLPGB PatchedLinux 6.13 Git20406080100SE +/- 0.11, N = 3SE +/- 0.24, N = 390.7890.42

Memcached

Set To Get Ratio: 1:5

OpenBenchmarking.orgOps/sec, More Is BetterMemcached 1.6.19Set To Get Ratio: 1:5INVLPGB PatchedLinux 6.13 Git800K1600K2400K3200K4000KSE +/- 7706.14, N = 3SE +/- 2641.20, N = 33583145.463569817.661. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre

Blender

Blend File: Barbershop - Compute: CPU-Only

OpenBenchmarking.orgSeconds, Fewer Is BetterBlender 4.3Blend File: Barbershop - Compute: CPU-OnlyINVLPGB PatchedLinux 6.13 Git306090120150SE +/- 0.12, N = 3SE +/- 0.13, N = 3124.48124.03

NAS Parallel Benchmarks

Test / Class: LU.C

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: LU.CINVLPGB PatchedLinux 6.13 Git70K140K210K280K350KSE +/- 2363.19, N = 10SE +/- 2270.42, N = 15311722.61312554.831. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz 2. Open MPI 4.1.6

Timed LLVM Compilation

Build System: Unix Makefiles

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed LLVM Compilation 16.0Build System: Unix MakefilesINVLPGB PatchedLinux 6.13 Git4080120160200SE +/- 0.59, N = 3SE +/- 0.32, N = 3161.06160.67

Apache IoTDB

Device Count: 500 - Batch Size Per Write: 100 - Sensor Count: 500 - Client Number: 400

OpenBenchmarking.orgAverage Latency, Fewer Is BetterApache IoTDB 1.2Device Count: 500 - Batch Size Per Write: 100 - Sensor Count: 500 - Client Number: 400INVLPGB PatchedLinux 6.13 Git4080120160200SE +/- 0.71, N = 3SE +/- 0.88, N = 3167.95168.36MAX: 26468.05MAX: 26450.24

NAS Parallel Benchmarks

Test / Class: IS.D

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: IS.DINVLPGB PatchedLinux 6.13 Git15003000450060007500SE +/- 68.35, N = 5SE +/- 60.97, N = 36984.657001.421. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz 2. Open MPI 4.1.6

Timed Godot Game Engine Compilation

Time To Compile

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Godot Game Engine Compilation 4.0Time To CompileINVLPGB PatchedLinux 6.13 Git20406080100SE +/- 0.16, N = 3SE +/- 0.05, N = 380.8380.69

NAMD

Input: STMV with 1,066,628 Atoms

OpenBenchmarking.orgns/day, More Is BetterNAMD 3.0Input: STMV with 1,066,628 AtomsINVLPGB PatchedLinux 6.13 Git0.94841.89682.84523.79364.742SE +/- 0.00334, N = 3SE +/- 0.00531, N = 34.209734.21510

DaCapo Benchmark

Java Test: Apache Kafka

OpenBenchmarking.orgmsec, Fewer Is BetterDaCapo Benchmark 23.11Java Test: Apache KafkaINVLPGB PatchedLinux 6.13 Git11002200330044005500SE +/- 6.08, N = 3SE +/- 4.26, N = 350505046

DaCapo Benchmark

Java Test: jMonkeyEngine

OpenBenchmarking.orgmsec, Fewer Is BetterDaCapo Benchmark 23.11Java Test: jMonkeyEngineINVLPGB PatchedLinux 6.13 Git15003000450060007500SE +/- 1.67, N = 3SE +/- 1.67, N = 368066808

Renaissance

Test: In-Memory Database Shootout

OpenBenchmarking.orgms, Fewer Is BetterRenaissance 0.16Test: In-Memory Database ShootoutINVLPGB PatchedLinux 6.13 Git10002000300040005000SE +/- 34.36, N = 3SE +/- 55.34, N = 44512.94511.6MIN: 4347.69 / MAX: 5072.64MIN: 4295.26 / MAX: 5024.51

NAMD

Input: ATPase with 327,506 Atoms

OpenBenchmarking.orgns/day, More Is BetterNAMD 3.0Input: ATPase with 327,506 AtomsINVLPGB PatchedLinux 6.13 Git3691215SE +/- 0.03, N = 3SE +/- 0.03, N = 312.2212.22

Quicksilver

Input: CORAL2 P2

OpenBenchmarking.orgFigure Of Merit, More Is BetterQuicksilver 20230818Input: CORAL2 P2INVLPGB PatchedLinux 6.13 Git5M10M15M20M25MSE +/- 18559.21, N = 3SE +/- 12018.50, N = 322056667220566671. (CXX) g++ options: -fopenmp -O3 -march=native

Xmrig

Variant: GhostRider - Hash Count: 1M

OpenBenchmarking.orgH/s, More Is BetterXmrig 6.21Variant: GhostRider - Hash Count: 1MINVLPGB PatchedLinux 6.13 Git3K6K9K12K15KSE +/- 32.37, N = 3SE +/- 668.22, N = 1516061.413706.91. (CXX) g++ options: -fexceptions -fno-rtti -maes -O3 -Ofast -static-libgcc -static-libstdc++ -rdynamic -lssl -lcrypto -luv -lpthread -lrt -ldl -lhwloc

RELION

Test: Basic - Device: CPU

OpenBenchmarking.orgSeconds, Fewer Is BetterRELION 5.0Test: Basic - Device: CPUINVLPGB PatchedLinux 6.13 Git50100150200250SE +/- 4.55, N = 12SE +/- 3.72, N = 12211.38214.441. (CXX) g++ options: -fPIC -std=c++14 -fopenmp -O3 -rdynamic -ldl -ltiff -lfftw3f -lfftw3 -lpng -ljpeg -lmpi_cxx -lmpi

NAS Parallel Benchmarks

Test / Class: MG.C

OpenBenchmarking.orgTotal Mop/s, More Is BetterNAS Parallel Benchmarks 3.4Test / Class: MG.CINVLPGB PatchedLinux 6.13 Git30K60K90K120K150KSE +/- 1764.21, N = 3SE +/- 2956.05, N = 15147724.66145177.791. (F9X) gfortran options: -O3 -march=native -lmpi_usempif08 -lmpi_mpifh -lmpi -lopen-rte -lopen-pal -lhwloc -levent_core -levent_pthreads -lm -lz 2. Open MPI 4.1.6


Phoronix Test Suite v10.8.5