Intel Raptor Lake Mitigation Cost

Benchmarks by Michael Larabel for a future article.

Compare your own system(s) to this result file with the Phoronix Test Suite by running the command: phoronix-test-suite benchmark 2211306-NE-INTELRAPT80
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

Multi-Way Comparison

Condense Multi-Option Tests Into Single Result Graphs

Table

Show Detailed System Result Table

Run Management

Highlight
Result
Toggle/Hide
Result
Result
Identifier
View Logs
Performance Per
Dollar
Date
Run
  Test
  Duration
Linux 6.1 Defaults
November 28 2022
  12 Hours, 40 Minutes
mitigations=off
November 29 2022
  12 Hours, 56 Minutes
Invert Behavior (Only Show Selected Data)
  12 Hours, 48 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):

Linux 6.1 Defaults

Kernel Notes: Transparent Huge Pages: madvise
Compiler Notes: --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,brig,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-link-serialization=2 --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-11-xKiWfi/gcc-11-11.3.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-11-xKiWfi/gcc-11-11.3.0/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
Processor Notes: Scaling Governor: intel_pstate powersave (EPP: balance_performance) - CPU Microcode: 0x10e - Thermald 2.4.9
Java Notes: OpenJDK Runtime Environment (build 11.0.17+8-post-Ubuntu-1ubuntu222.04)
Python Notes: Python 3.10.6
Security Notes: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: 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 IBRS IBPB: conditional RSB filling PBRSB-eIBRS: SW sequence + srbds: Not affected + tsx_async_abort: Not affected

mitigations=off

Processor: Intel Core i9-13900K @ 5.50GHz (24 Cores / 32 Threads), Motherboard: ASUS PRIME Z790-P WIFI (0602 BIOS), Chipset: Intel Device 7a27, Memory: 32GB, Disk: 1000GB Western Digital WDS100T1X0E-00AFY0, Graphics: AMD Radeon RX 6800 16GB (2475/1000MHz), Audio: Realtek ALC897, Monitor: ASUS VP28U, Network: Realtek RTL8125 2.5GbE + Intel Device 7a70

OS: Ubuntu 22.04, Kernel: 6.1.0-060100rc5daily20221117-generic (x86_64), Desktop: GNOME Shell 42.5, Display Server: X Server 1.21.1.3 + Wayland, OpenGL: 4.6 Mesa 22.0.5 (LLVM 13.0.1 DRM 3.49), Vulkan: 1.3.204, Compiler: GCC 11.3.0, File-System: ext4, Screen Resolution: 3840x2160

Kernel Notes: Transparent Huge Pages: madvise
Compiler Notes: --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,brig,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-link-serialization=2 --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-11-xKiWfi/gcc-11-11.3.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-11-xKiWfi/gcc-11-11.3.0/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
Processor Notes: Scaling Governor: intel_pstate powersave (EPP: balance_performance) - CPU Microcode: 0x10e - Thermald 2.4.9
Java Notes: OpenJDK Runtime Environment (build 11.0.17+8-post-Ubuntu-1ubuntu222.04)
Python Notes: Python 3.10.6
Security Notes: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Not affected + spec_store_bypass: Vulnerable + spectre_v1: Vulnerable: __user pointer sanitization and usercopy barriers only; no swapgs barriers + spectre_v2: Vulnerable IBPB: disabled STIBP: disabled PBRSB-eIBRS: Vulnerable + srbds: Not affected + tsx_async_abort: Not affected

Intel Raptor Lake Mitigation CostOpenBenchmarking.orgPhoronix Test SuiteIntel Core i9-13900K @ 5.50GHz (24 Cores / 32 Threads)ASUS PRIME Z790-P WIFI (0602 BIOS)Intel Device 7a2732GB1000GB Western Digital WDS100T1X0E-00AFY0AMD Radeon RX 6800 16GB (2475/1000MHz)Realtek ALC897ASUS VP28URealtek RTL8125 2.5GbE + Intel Device 7a70Ubuntu 22.046.1.0-060100rc5daily20221117-generic (x86_64)GNOME Shell 42.5X Server 1.21.1.3 + Wayland4.6 Mesa 22.0.5 (LLVM 13.0.1 DRM 3.49)1.3.204GCC 11.3.0ext43840x2160ProcessorMotherboardChipsetMemoryDiskGraphicsAudioMonitorNetworkOSKernelDesktopDisplay ServerOpenGLVulkanCompilerFile-SystemScreen ResolutionIntel Raptor Lake Mitigation Cost 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,brig,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-link-serialization=2 --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-11-xKiWfi/gcc-11-11.3.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-11-xKiWfi/gcc-11-11.3.0/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: intel_pstate powersave (EPP: balance_performance) - CPU Microcode: 0x10e - Thermald 2.4.9 - OpenJDK Runtime Environment (build 11.0.17+8-post-Ubuntu-1ubuntu222.04)- Python 3.10.6- Linux 6.1 Defaults: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: 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 IBRS IBPB: conditional RSB filling PBRSB-eIBRS: SW sequence + srbds: Not affected + tsx_async_abort: Not affected - mitigations=off: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Not affected + spec_store_bypass: Vulnerable + spectre_v1: Vulnerable: __user pointer sanitization and usercopy barriers only; no swapgs barriers + spectre_v2: Vulnerable IBPB: disabled STIBP: disabled PBRSB-eIBRS: Vulnerable + srbds: Not affected + tsx_async_abort: Not affected

Linux 6.1 Defaults vs. mitigations=off ComparisonPhoronix Test SuiteBaseline+9.4%+9.4%+18.8%+18.8%+28.2%+28.2%37.4%9%8.8%7.9%6.9%5.6%4.8%3.8%3.8%3.8%3.5%3.2%3%2.9%2.8%2.5%2.4%2.3%2%Context SwitchingScala Dotty1000000 - Rand Read - 11000000 - 100 - B.I.J.T.T1000000 - 100 - I.J.T.T1000000 - Rand Read - 1Mutex1000000 - Rand Read - 16Q.1.C.E.5300 - 1000 - 30Socket ActivityRead While Writing3.2%C.S.T3.2%Memory Allocations1000000 - Rand Read - 1610, LosslessFutex2.8%Speed 9 Realtime - Bosphorus 4K6, Lossless1000000 - Increment - 16Speed 0 Two-Pass - Bosphorus 4K2.4%I.M.D.S2.4%A.U.C.T2.4%MMAP2.4%4.P.1.P.S.6.Q2.4%Speed 10 Realtime - Bosphorus 4K100 - 1000 - 302.2%4.P.1.P.S.6.Q2.1%Apache Spark ALSAtomic2%Stress-NGRenaissanceApache HBaseApache SparkApache SparkApache HBaseStress-NGApache HBaseWebP2 Image EncodeApache CouchDBStress-NGFacebook RocksDBctx_clockOSBenchApache HBaselibavif avifencStress-NGAOM AV1libavif avifencApache HBaseAOM AV1RenaissanceRenaissanceStress-NGsrsRANAOM AV1Apache CouchDBsrsRANRenaissanceStress-NGLinux 6.1 Defaultsmitigations=off

Intel Raptor Lake Mitigation Costrenaissance: Scala Dottyhbase: 1000000 - Rand Read - 1hbase: 1000000 - Rand Read - 1hbase: 1000000 - Rand Read - 16couchdb: 300 - 1000 - 30stress-ng: Socket Activityctx-clock: Context Switch Timeosbench: Memory Allocationshbase: 1000000 - Rand Read - 16avifenc: 10, Losslessstress-ng: Futexaom-av1: Speed 9 Realtime - Bosphorus 4Kavifenc: 6, Losslesshbase: 1000000 - Increment - 16aom-av1: Speed 0 Two-Pass - Bosphorus 4Krenaissance: In-Memory Database Shootoutrenaissance: Akka Unbalanced Cobwebbed Treestress-ng: MMAPsrsran: 4G PHY_DL_Test 100 PRB SISO 64-QAMaom-av1: Speed 10 Realtime - Bosphorus 4Kcouchdb: 100 - 1000 - 30srsran: 4G PHY_DL_Test 100 PRB SISO 64-QAMrenaissance: Apache Spark ALSspark: 1000000 - 100 - Repartition Test Timeavifenc: 6renaissance: Genetic Algorithm Using Jenetics + Futureswebp2: Defaulthbase: 1000000 - Increment - 16osbench: Create Processesrenaissance: Finagle HTTP Requestsdragonflydb: 50 - 1:1clickhouse: 100M Rows Web Analytics Dataset, Third Runstress-ng: SENDFILErawtherapee: Total Benchmark Timedragonflydb: 50 - 5:1stress-ng: NUMAsvt-av1: Preset 8 - Bosphorus 4Kdeepsparse: NLP Text Classification, BERT base uncased SST2 - Asynchronous Multi-Streamdeepsparse: NLP Text Classification, BERT base uncased SST2 - Asynchronous Multi-Streamrenaissance: Apache Spark PageRankstress-ng: System V Message Passingsvt-av1: Preset 10 - Bosphorus 4Kspark: 1000000 - 100 - Calculate Pi Benchmark Using Dataframeaom-av1: Speed 6 Realtime - Bosphorus 4Ksrsran: 5G PHY_DL_NR Test 52 PRB SISO 64-QAMstress-ng: Forkingdeepsparse: NLP Document Classification, oBERT base uncased on IMDB - Asynchronous Multi-Streamastcenc: Thoroughaom-av1: Speed 8 Realtime - Bosphorus 4Kwebp: Quality 100, Lossless, Highest Compressionhbase: 1000000 - Async Rand Write - 16deepsparse: NLP Document Classification, oBERT base uncased on IMDB - Asynchronous Multi-Streamclickhouse: 100M Rows Web Analytics Dataset, Second Runhbase: 1000000 - Async Rand Write - 16stress-ng: Glibc C String Functionsopenvino: Person Vehicle Bike Detection FP16 - CPUopenvino: Person Vehicle Bike Detection FP16 - CPUspark: 1000000 - 500 - Repartition Test Timeosbench: Create Filesstress-ng: Matrix Mathdacapobench: Tradesoaprocksdb: Rand Fill Syncrocksdb: Rand Fillsrsran: 4G PHY_DL_Test 100 PRB MIMO 256-QAMmysqlslap: 32srsran: 5G PHY_DL_NR Test 52 PRB SISO 64-QAMkeydb: rocksdb: Update Randcouchdb: 100 - 3000 - 30renaissance: Savina Reactors.IOdragonflydb: 200 - 1:1openvino: Weld Porosity Detection FP16-INT8 - CPUopenvino: Weld Porosity Detection FP16-INT8 - CPUstress-ng: IO_uringastcenc: Exhaustivesrsran: 4G PHY_DL_Test 100 PRB MIMO 64-QAMpgbench: 100 - 50 - Read Onlypgbench: 100 - 100 - Read Onlyinfluxdb: 4 - 10000 - 2,5000,1 - 10000srsran: 4G PHY_DL_Test 100 PRB MIMO 64-QAMstress-ng: Glibc Qsort Data Sortingdeepsparse: CV Detection,YOLOv5s COCO - Asynchronous Multi-Streamspark: 1000000 - 500 - Calculate Pi Benchmark Using Dataframedacapobench: Tradebeansdeepsparse: NLP Text Classification, DistilBERT mnli - Asynchronous Multi-Streambork: File Encryption Timerenaissance: Apache Spark Bayesopenvino: Person Detection FP32 - CPUopenvino: Person Detection FP32 - CPUdeepsparse: CV Detection,YOLOv5s COCO - Asynchronous Multi-Streamopenvino: Vehicle Detection FP16 - CPUdeepsparse: NLP Token Classification, BERT base uncased conll2003 - Asynchronous Multi-Streamrocksdb: Seq Fillopenvino: Vehicle Detection FP16 - CPUdeepsparse: NLP Text Classification, DistilBERT mnli - Asynchronous Multi-Streamsvt-av1: Preset 4 - Bosphorus 4Kopenvino: Vehicle Detection FP16-INT8 - CPUdeepsparse: NLP Token Classification, BERT base uncased conll2003 - Asynchronous Multi-Streamdeepsparse: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Asynchronous Multi-Streamopenvino: Vehicle Detection FP16-INT8 - CPUrocksdb: Read Rand Write Randosbench: Launch Programsavifenc: 0avifenc: 2svt-av1: Preset 12 - Bosphorus 4Kstargate: 192000 - 1024cryptsetup: AES-XTS 256b Encryptionsrsran: OFDM_Testaircrack-ng: stress-ng: Memory Copyingstress-ng: Cryptocryptsetup: AES-XTS 256b Decryptiondacapobench: Jythondragonflydb: 50 - 1:5aom-av1: Speed 4 Two-Pass - Bosphorus 4Krenaissance: Rand Forestclickhouse: 100M Rows Web Analytics Dataset, First Run / Cold Cachewebp: Quality 100deepsparse: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Asynchronous Multi-Streamopenvino: Person Detection FP16 - CPUspark: 1000000 - 500 - Calculate Pi Benchmarksqlite-speedtest: Timed Time - Size 1,000nginx: 100stargate: 192000 - 512cryptsetup: AES-XTS 512b Encryptionwebp: Defaultjava-gradle-perf: Reactoropenvino: Weld Porosity Detection FP16 - CPUopenvino: Weld Porosity Detection FP16 - CPUnginx: 500cryptsetup: AES-XTS 512b Decryptionnginx: 1000cryptsetup: PBKDF2-whirlpoolsrsran: 4G PHY_DL_Test 100 PRB SISO 256-QAMopenvino: Face Detection FP16-INT8 - CPUinfluxdb: 64 - 10000 - 2,5000,1 - 10000build-gem5: Time To Compilecryptsetup: Serpent-XTS 512b Encryptionopenvino: Machine Translation EN To DE FP16 - CPUopenvino: Person Detection FP16 - CPUstress-ng: Vector Mathnginx: 200openvino: Face Detection FP16-INT8 - CPUosbench: Create Threadsbuild-linux-kernel: defconfigstress-ng: CPU Stressastcenc: Fastsrsran: 4G PHY_DL_Test 100 PRB MIMO 256-QAMaom-av1: Speed 6 Two-Pass - Bosphorus 4Kbrl-cad: VGR Performance Metricastcenc: Mediumcouchdb: 300 - 3000 - 30build-linux-kernel: allmodconfigcryptsetup: Serpent-XTS 256b Decryptionopenvino: Machine Translation EN To DE FP16 - CPUspark: 1000000 - 100 - Calculate Pi Benchmarkdragonflydb: 200 - 5:1openvino: Face Detection FP16 - CPUopenvino: Age Gender Recognition Retail 0013 FP16 - CPUstress-ng: Semaphorescryptsetup: PBKDF2-sha512cryptsetup: Twofish-XTS 256b Decryptiondeepsparse: CV Classification, ResNet-50 ImageNet - Asynchronous Multi-Streamopenvino: Age Gender Recognition Retail 0013 FP16-INT8 - CPUcryptsetup: Serpent-XTS 512b Decryptioncryptsetup: Twofish-XTS 512b Encryptionstress-ng: Mallocstress-ng: x86_64 RdRandcryptsetup: Twofish-XTS 256b Encryptioncryptsetup: Twofish-XTS 512b Decryptionrocksdb: Rand Readdeepsparse: CV Classification, ResNet-50 ImageNet - Asynchronous Multi-Streamcryptsetup: Serpent-XTS 256b Encryptionrenaissance: ALS Movie Lensdragonflydb: 200 - 1:5openvino: Age Gender Recognition Retail 0013 FP16-INT8 - CPUopenvino: Age Gender Recognition Retail 0013 FP16 - CPUopenvino: Face Detection FP16 - CPUpgbench: 100 - 100 - Read Only - Average Latencypgbench: 100 - 50 - Read Only - Average Latencyspark: 1000000 - 100 - Group By Test Timespark: 1000000 - 100 - SHA-512 Benchmark Timesrsran: 4G PHY_DL_Test 100 PRB SISO 256-QAMwebp2: Quality 100, Lossless Compressionwebp2: Quality 95, Compression Effort 7webp2: Quality 75, Compression Effort 7webp: Quality 100, Highest Compressionwebp: Quality 100, Losslessrocksdb: Read While Writingstress-ng: Context Switchingstress-ng: CPU Cachestress-ng: Atomicstress-ng: Mutexspark: 1000000 - 500 - Broadcast Inner Join Test Timespark: 1000000 - 100 - Broadcast Inner Join Test Timespark: 1000000 - 100 - Inner Join Test Timewebp2: Quality 100, Compression Effort 5dacapobench: H2Linux 6.1 Defaultsmitigations=off451.425419385494.25325195.7712442.1269742911173.4243515710.9183.605.5891252080.421921.47188.4664.95646.284.7153.849229.11592.71.053.2991033.214.9212617.2026951992.96335702.64309.00610692.1031.6066159328.58688.8076.345220.946454.16461900.632462582.17146.0923.3444.19104.3132107.51948.177616.240265.780.937878312.5442303.652023741573.78755.3610.581.088.312671102476.691642389271356117706.43304227.51507777.21770017168.6614188.36696607.2814.411664.6427773.351.4516643.2126088512306882696816.3181.5414.3775.64433.381709112.77034.845701.82223.373.58158.0568372.54947.6739150243121.45106.26593.059919.0412.508350.53358.70364115523.99285670.34434.257216.7833.0335337107.719983333369682.1437302.7442801.847108.817096742394.0112.09380.0300.0416.69236.76313.5951.2931.307207280.772.6355906523.525.88141.722473.8150.57203178.406503.1187512.541334634245.1437.502980179.4174.6401011.763.952212.51134428.58211451.2918.269.11633235.65251499.15345.1054202.021.70462945126.1469287.567410.7031043.6125.0051.216440950.321556.5414690.903517936.222839109704.6155.273233202.881043.4699.632335069.7682763.69698.4704.217748085377.18681009.04371.97029354.710.721.635.130.0810.0402.551.99713.60.040.160.335.182.34508988711794492.4397.76338549.5616677300.350.790.820.939.112111414.327660365290.81226073.1412840.8286252997083.3263421101.2885.915.4531282640.411968.27359.2649.59631.386.6255.017224.31561.11.073.2421051.315.1712417.4705191963.96423833.88304.77602354.7931.1786243118.04679.6677.344223.813553.48561877.832855914.63147.8503.3844.71103.1133593.49937.654016.063666.490.947962912.6734306.762003778821.91748.0610.681.078.239698101602.631656386071367267712.13330225.81518638.23764558167.4744159.16650409.6014.511653.3627589.431.4612639.1126865712382652680364.8180.4411.9175.19753.401719113.42714.817697.82236.033.56158.9123370.54952.7514149461021.56105.73463.044914.6212.450350.30078.74362464423.88715770.03534.107217.7343.0467167078.420060000069948.1877275.1442952.327083.917156765462.4212.05378.8299.1116.64237.43823.6051.4231.385207784.832.6292256508.025.82141.421472.8150.67202777.496490.6187868.731336901245.5438.112976169.8174.4101010.463.872209.76134263.01211212.3718.249.10679535.68851447.48344.7621201.821.68463365126.2613287.310410.3421042.7125.1051.256436491.371557.3614683.753519548.922840387704.3155.323833193.111043.1699.832341790.8582746.87698.5704.117750312277.19621008.94371.57029304.50.721.635.130.0810.0402.551.99713.60.040.160.335.182.34493058116210513.9799.42331842.5917479306.270.780.760.879.462137OpenBenchmarking.org

Renaissance

Renaissance is a suite of benchmarks designed to test the Java JVM from Apache Spark to a Twitter-like service to Scala and other features. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterRenaissance 0.14Test: Scala Dottymitigations=offLinux 6.1 Defaults100200300400500SE +/- 0.49, N = 3SE +/- 6.25, N = 15414.3451.4MIN: 348.11 / MAX: 744.04MIN: 343.36 / MAX: 782.31

Apache HBase

This is a benchmark of the Apache HBase non-relational distributed database system inspired from Google's Bigtable. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgRows Per Second, More Is BetterApache HBase 2.5.0Rows: 1000000 - Test: Random Read - Clients: 1mitigations=offLinux 6.1 Defaults6K12K18K24K30KSE +/- 87.43, N = 3SE +/- 201.55, N = 32766025419

OpenBenchmarking.orgMicroseconds - Average Latency, Fewer Is BetterApache HBase 2.5.0Rows: 1000000 - Test: Random Read - Clients: 1mitigations=offLinux 6.1 Defaults918273645SE +/- 0.33, N = 3SE +/- 0.33, N = 33638

OpenBenchmarking.orgMicroseconds - Average Latency, Fewer Is BetterApache HBase 2.5.0Rows: 1000000 - Test: Random Read - Clients: 16mitigations=offLinux 6.1 Defaults1224364860SE +/- 0.33, N = 3SE +/- 0.58, N = 35254

Apache CouchDB

This is a bulk insertion benchmark of Apache CouchDB. CouchDB is a document-oriented NoSQL database implemented in Erlang. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterApache CouchDB 3.2.2Bulk Size: 300 - Inserts: 1000 - Rounds: 30mitigations=offLinux 6.1 Defaults20406080100SE +/- 0.99, N = 4SE +/- 1.45, N = 1590.8194.251. (CXX) g++ options: -std=c++17 -lmozjs-91 -lm -lei -fPIC -MMD

Stress-NG

Stress-NG is a Linux stress tool developed by Colin King of Canonical. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: Socket Activitymitigations=offLinux 6.1 Defaults6K12K18K24K30KSE +/- 323.81, N = 15SE +/- 197.02, N = 326073.1425195.771. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lrt -lsctp -lz -pthread

ctx_clock

Ctx_clock is a simple test program to measure the context switch time in clock cycles. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgClocks, Fewer Is Betterctx_clockContext Switch Timemitigations=offLinux 6.1 Defaults306090120150SE +/- 0.95, N = 11SE +/- 0.00, N = 3128124

OSBench

OSBench is a collection of micro-benchmarks for measuring operating system primitives like time to create threads/processes, launching programs, creating files, and memory allocation. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgNs Per Event, Fewer Is BetterOSBenchTest: Memory Allocationsmitigations=offLinux 6.1 Defaults1020304050SE +/- 0.05, N = 3SE +/- 0.11, N = 340.8342.131. (CC) gcc options: -lm

Apache HBase

This is a benchmark of the Apache HBase non-relational distributed database system inspired from Google's Bigtable. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgRows Per Second, More Is BetterApache HBase 2.5.0Rows: 1000000 - Test: Random Read - Clients: 16mitigations=offLinux 6.1 Defaults60K120K180K240K300KSE +/- 984.52, N = 3SE +/- 2676.47, N = 3299708291117

libavif avifenc

This is a test of the AOMedia libavif library testing the encoding of a JPEG image to AV1 Image Format (AVIF). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.11Encoder Speed: 10, Losslessmitigations=offLinux 6.1 Defaults0.77041.54082.31123.08163.852SE +/- 0.033, N = 3SE +/- 0.031, N = 73.3263.4241. (CXX) g++ options: -O3 -fPIC -lm

Stress-NG

Stress-NG is a Linux stress tool developed by Colin King of Canonical. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: Futexmitigations=offLinux 6.1 Defaults800K1600K2400K3200K4000KSE +/- 41423.32, N = 15SE +/- 41459.49, N = 153421101.283515710.911. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lrt -lsctp -lz -pthread

AOM AV1

This is a test of the AOMedia AV1 encoder (libaom) developed by AOMedia and Google as the AV1 Codec Library. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.5Encoder Mode: Speed 9 Realtime - Input: Bosphorus 4Kmitigations=offLinux 6.1 Defaults20406080100SE +/- 0.19, N = 3SE +/- 0.19, N = 385.9183.601. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm

libavif avifenc

This is a test of the AOMedia libavif library testing the encoding of a JPEG image to AV1 Image Format (AVIF). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.11Encoder Speed: 6, Losslessmitigations=offLinux 6.1 Defaults1.25752.5153.77255.036.2875SE +/- 0.051, N = 6SE +/- 0.060, N = 35.4535.5891. (CXX) g++ options: -O3 -fPIC -lm

Apache HBase

This is a benchmark of the Apache HBase non-relational distributed database system inspired from Google's Bigtable. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgRows Per Second, More Is BetterApache HBase 2.5.0Rows: 1000000 - Test: Increment - Clients: 16mitigations=offLinux 6.1 Defaults30K60K90K120K150KSE +/- 1763.01, N = 3SE +/- 1182.66, N = 3128264125208

AOM AV1

This is a test of the AOMedia AV1 encoder (libaom) developed by AOMedia and Google as the AV1 Codec Library. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.5Encoder Mode: Speed 0 Two-Pass - Input: Bosphorus 4Kmitigations=offLinux 6.1 Defaults0.09450.1890.28350.3780.4725SE +/- 0.00, N = 3SE +/- 0.00, N = 30.410.421. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm

Renaissance

Renaissance is a suite of benchmarks designed to test the Java JVM from Apache Spark to a Twitter-like service to Scala and other features. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterRenaissance 0.14Test: In-Memory Database Shootoutmitigations=offLinux 6.1 Defaults400800120016002000SE +/- 24.16, N = 3SE +/- 27.32, N = 31968.21921.4MIN: 1760.93 / MAX: 2449.66MIN: 1708.07 / MAX: 2253.56

OpenBenchmarking.orgms, Fewer Is BetterRenaissance 0.14Test: Akka Unbalanced Cobwebbed Treemitigations=offLinux 6.1 Defaults16003200480064008000SE +/- 34.79, N = 3SE +/- 19.11, N = 37359.27188.4MIN: 5648.82 / MAX: 7428.65MIN: 5401.58 / MAX: 7219.72

Stress-NG

Stress-NG is a Linux stress tool developed by Colin King of Canonical. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: MMAPmitigations=offLinux 6.1 Defaults140280420560700SE +/- 7.79, N = 3SE +/- 6.66, N = 3649.59664.951. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lrt -lsctp -lz -pthread

srsRAN

srsRAN is an open-source LTE/5G software radio suite created by Software Radio Systems (SRS). The srsRAN radio suite was formerly known as srsLTE and can be used for building your own software-defined radio (SDR) 4G/5G mobile network. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgeNb Mb/s, More Is BettersrsRAN 22.04.1Test: 4G PHY_DL_Test 100 PRB SISO 64-QAMmitigations=offLinux 6.1 Defaults140280420560700SE +/- 6.50, N = 5SE +/- 2.79, N = 3631.3646.21. (CXX) g++ options: -std=c++14 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -ldl -lpthread -lm

AOM AV1

This is a test of the AOMedia AV1 encoder (libaom) developed by AOMedia and Google as the AV1 Codec Library. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.5Encoder Mode: Speed 10 Realtime - Input: Bosphorus 4Kmitigations=offLinux 6.1 Defaults20406080100SE +/- 0.57, N = 3SE +/- 0.27, N = 386.6284.711. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm

Apache CouchDB

This is a bulk insertion benchmark of Apache CouchDB. CouchDB is a document-oriented NoSQL database implemented in Erlang. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterApache CouchDB 3.2.2Bulk Size: 100 - Inserts: 1000 - Rounds: 30mitigations=offLinux 6.1 Defaults1224364860SE +/- 0.45, N = 3SE +/- 0.07, N = 355.0253.851. (CXX) g++ options: -std=c++17 -lmozjs-91 -lm -lei -fPIC -MMD

srsRAN

srsRAN is an open-source LTE/5G software radio suite created by Software Radio Systems (SRS). The srsRAN radio suite was formerly known as srsLTE and can be used for building your own software-defined radio (SDR) 4G/5G mobile network. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgUE Mb/s, More Is BettersrsRAN 22.04.1Test: 4G PHY_DL_Test 100 PRB SISO 64-QAMmitigations=offLinux 6.1 Defaults50100150200250SE +/- 1.80, N = 5SE +/- 0.44, N = 3224.3229.11. (CXX) g++ options: -std=c++14 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -ldl -lpthread -lm

Renaissance

Renaissance is a suite of benchmarks designed to test the Java JVM from Apache Spark to a Twitter-like service to Scala and other features. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterRenaissance 0.14Test: Apache Spark ALSmitigations=offLinux 6.1 Defaults30060090012001500SE +/- 8.72, N = 3SE +/- 4.17, N = 31561.11592.7MIN: 1476.82 / MAX: 1681.97MIN: 1505.66 / MAX: 1729.7

Apache Spark

This is a benchmark of Apache Spark with its PySpark interface. Apache Spark is an open-source unified analytics engine for large-scale data processing and dealing with big data. This test profile benchmars the Apache Spark in a single-system configuration using spark-submit. The test makes use of DIYBigData's pyspark-benchmark (https://github.com/DIYBigData/pyspark-benchmark/) for generating of test data and various Apache Spark operations. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterApache Spark 3.3Row Count: 1000000 - Partitions: 100 - Repartition Test Timemitigations=offLinux 6.1 Defaults0.24080.48160.72240.96321.204SE +/- 0.01, N = 15SE +/- 0.01, N = 151.071.05

libavif avifenc

This is a test of the AOMedia libavif library testing the encoding of a JPEG image to AV1 Image Format (AVIF). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.11Encoder Speed: 6mitigations=offLinux 6.1 Defaults0.74231.48462.22692.96923.7115SE +/- 0.011, N = 3SE +/- 0.034, N = 43.2423.2991. (CXX) g++ options: -O3 -fPIC -lm

Renaissance

Renaissance is a suite of benchmarks designed to test the Java JVM from Apache Spark to a Twitter-like service to Scala and other features. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterRenaissance 0.14Test: Genetic Algorithm Using Jenetics + Futuresmitigations=offLinux 6.1 Defaults2004006008001000SE +/- 7.18, N = 3SE +/- 12.61, N = 41051.31033.2MIN: 988.02 / MAX: 1106.1MIN: 963.25 / MAX: 1123.84

WebP2 Image Encode

This is a test of Google's libwebp2 library with the WebP2 image encode utility and using a sample 6000x4000 pixel JPEG image as the input, similar to the WebP/libwebp test profile. WebP2 is currently experimental and under heavy development as ultimately the successor to WebP. WebP2 supports 10-bit HDR, more efficienct lossy compression, improved lossless compression, animation support, and full multi-threading support compared to WebP. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMP/s, More Is BetterWebP2 Image Encode 20220823Encode Settings: Defaultmitigations=offLinux 6.1 Defaults48121620SE +/- 0.14, N = 7SE +/- 0.20, N = 315.1714.921. (CXX) g++ options: -msse4.2 -fno-rtti -O3 -ldl

Apache HBase

This is a benchmark of the Apache HBase non-relational distributed database system inspired from Google's Bigtable. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMicroseconds - Average Latency, Fewer Is BetterApache HBase 2.5.0Rows: 1000000 - Test: Increment - Clients: 16mitigations=offLinux 6.1 Defaults306090120150SE +/- 1.67, N = 3SE +/- 1.33, N = 3124126

OSBench

OSBench is a collection of micro-benchmarks for measuring operating system primitives like time to create threads/processes, launching programs, creating files, and memory allocation. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgus Per Event, Fewer Is BetterOSBenchTest: Create Processesmitigations=offLinux 6.1 Defaults48121620SE +/- 0.17, N = 3SE +/- 0.06, N = 317.4717.201. (CC) gcc options: -lm

Renaissance

Renaissance is a suite of benchmarks designed to test the Java JVM from Apache Spark to a Twitter-like service to Scala and other features. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterRenaissance 0.14Test: Finagle HTTP Requestsmitigations=offLinux 6.1 Defaults400800120016002000SE +/- 12.55, N = 3SE +/- 2.14, N = 31963.91992.9MIN: 1798.97 / MAX: 2151.27MIN: 1809.15 / MAX: 2148.58

Dragonflydb

Dragonfly is an open-source database server that is a "modern Redis replacement" that aims to be the fastest memory store while being compliant with the Redis and Memcached protocols. For benchmarking Dragonfly, Memtier_benchmark is used as a NoSQL Redis/Memcache traffic generation plus benchmarking tool developed by Redis Labs. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgOps/sec, More Is BetterDragonflydb 0.6Clients: 50 - Set To Get Ratio: 1:1mitigations=offLinux 6.1 Defaults1.4M2.8M4.2M5.6M7MSE +/- 36510.22, N = 3SE +/- 54773.05, N = 36423833.886335702.641. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre

ClickHouse

ClickHouse is an open-source, high performance OLAP data management system. This test profile uses ClickHouse's standard benchmark recommendations per https://clickhouse.com/docs/en/operations/performance-test/ with the 100 million rows web analytics dataset. The reported value is the query processing time using the geometric mean of all queries performed. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgQueries Per Minute, Geo Mean, More Is BetterClickHouse 22.5.4.19100M Rows Web Analytics Dataset, Third Runmitigations=offLinux 6.1 Defaults70140210280350SE +/- 2.73, N = 15SE +/- 1.24, N = 15304.77309.00MIN: 23.27 / MAX: 30000MIN: 24.35 / MAX: 300001. ClickHouse server version 22.5.4.19 (official build).

Stress-NG

Stress-NG is a Linux stress tool developed by Colin King of Canonical. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: SENDFILEmitigations=offLinux 6.1 Defaults130K260K390K520K650KSE +/- 7176.09, N = 4SE +/- 7415.69, N = 3602354.79610692.101. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lrt -lsctp -lz -pthread

RawTherapee

RawTherapee is a cross-platform, open-source multi-threaded RAW image processing program. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterRawTherapeeTotal Benchmark Timemitigations=offLinux 6.1 Defaults714212835SE +/- 0.20, N = 3SE +/- 0.32, N = 331.1831.611. RawTherapee, version 5.8, command line.

Dragonflydb

Dragonfly is an open-source database server that is a "modern Redis replacement" that aims to be the fastest memory store while being compliant with the Redis and Memcached protocols. For benchmarking Dragonfly, Memtier_benchmark is used as a NoSQL Redis/Memcache traffic generation plus benchmarking tool developed by Redis Labs. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgOps/sec, More Is BetterDragonflydb 0.6Clients: 50 - Set To Get Ratio: 5:1mitigations=offLinux 6.1 Defaults1.3M2.6M3.9M5.2M6.5MSE +/- 5261.63, N = 3SE +/- 22475.30, N = 36243118.046159328.581. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre

Stress-NG

Stress-NG is a Linux stress tool developed by Colin King of Canonical. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: NUMAmitigations=offLinux 6.1 Defaults150300450600750SE +/- 3.21, N = 3SE +/- 9.66, N = 3679.66688.801. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lrt -lsctp -lz -pthread

SVT-AV1

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 1.2Encoder Mode: Preset 8 - Input: Bosphorus 4Kmitigations=offLinux 6.1 Defaults20406080100SE +/- 0.72, N = 7SE +/- 0.86, N = 477.3476.351. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq

Neural Magic DeepSparse

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.1Model: NLP Text Classification, BERT base uncased SST2 - Scenario: Asynchronous Multi-Streammitigations=offLinux 6.1 Defaults50100150200250SE +/- 0.57, N = 3SE +/- 0.42, N = 3223.81220.95

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.1Model: NLP Text Classification, BERT base uncased SST2 - Scenario: Asynchronous Multi-Streammitigations=offLinux 6.1 Defaults1224364860SE +/- 0.13, N = 3SE +/- 0.10, N = 353.4954.16

Renaissance

Renaissance is a suite of benchmarks designed to test the Java JVM from Apache Spark to a Twitter-like service to Scala and other features. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterRenaissance 0.14Test: Apache Spark PageRankmitigations=offLinux 6.1 Defaults400800120016002000SE +/- 7.79, N = 3SE +/- 15.25, N = 31877.81900.6MIN: 1729.93 / MAX: 1920.13MIN: 1751.32 / MAX: 1979.24

Stress-NG

Stress-NG is a Linux stress tool developed by Colin King of Canonical. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: System V Message Passingmitigations=offLinux 6.1 Defaults7M14M21M28M35MSE +/- 441148.10, N = 3SE +/- 12278.20, N = 332855914.6332462582.171. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lrt -lsctp -lz -pthread

SVT-AV1

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 1.2Encoder Mode: Preset 10 - Input: Bosphorus 4Kmitigations=offLinux 6.1 Defaults306090120150SE +/- 1.43, N = 3SE +/- 1.34, N = 3147.85146.091. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq

Apache Spark

This is a benchmark of Apache Spark with its PySpark interface. Apache Spark is an open-source unified analytics engine for large-scale data processing and dealing with big data. This test profile benchmars the Apache Spark in a single-system configuration using spark-submit. The test makes use of DIYBigData's pyspark-benchmark (https://github.com/DIYBigData/pyspark-benchmark/) for generating of test data and various Apache Spark operations. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterApache Spark 3.3Row Count: 1000000 - Partitions: 100 - Calculate Pi Benchmark Using Dataframemitigations=offLinux 6.1 Defaults0.76051.5212.28153.0423.8025SE +/- 0.04, N = 15SE +/- 0.05, N = 153.383.34

AOM AV1

This is a test of the AOMedia AV1 encoder (libaom) developed by AOMedia and Google as the AV1 Codec Library. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.5Encoder Mode: Speed 6 Realtime - Input: Bosphorus 4Kmitigations=offLinux 6.1 Defaults1020304050SE +/- 0.22, N = 3SE +/- 0.13, N = 344.7144.191. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm

srsRAN

srsRAN is an open-source LTE/5G software radio suite created by Software Radio Systems (SRS). The srsRAN radio suite was formerly known as srsLTE and can be used for building your own software-defined radio (SDR) 4G/5G mobile network. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgUE Mb/s, More Is BettersrsRAN 22.04.1Test: 5G PHY_DL_NR Test 52 PRB SISO 64-QAMmitigations=offLinux 6.1 Defaults20406080100SE +/- 0.52, N = 3SE +/- 0.12, N = 3103.1104.31. (CXX) g++ options: -std=c++14 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -ldl -lpthread -lm

Stress-NG

Stress-NG is a Linux stress tool developed by Colin King of Canonical. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: Forkingmitigations=offLinux 6.1 Defaults30K60K90K120K150KSE +/- 714.85, N = 3SE +/- 866.53, N = 3133593.49132107.511. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lrt -lsctp -lz -pthread

Neural Magic DeepSparse

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.1Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Asynchronous Multi-Streammitigations=offLinux 6.1 Defaults2004006008001000SE +/- 1.73, N = 3SE +/- 3.52, N = 3937.65948.18

ASTC Encoder

ASTC Encoder (astcenc) is for the Adaptive Scalable Texture Compression (ASTC) format commonly used with OpenGL, OpenGL ES, and Vulkan graphics APIs. This test profile does a coding test of both compression/decompression. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMT/s, More Is BetterASTC Encoder 4.0Preset: Thoroughmitigations=offLinux 6.1 Defaults48121620SE +/- 0.08, N = 3SE +/- 0.01, N = 316.0616.241. (CXX) g++ options: -O3 -flto -pthread

AOM AV1

This is a test of the AOMedia AV1 encoder (libaom) developed by AOMedia and Google as the AV1 Codec Library. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.5Encoder Mode: Speed 8 Realtime - Input: Bosphorus 4Kmitigations=offLinux 6.1 Defaults1530456075SE +/- 0.78, N = 3SE +/- 0.37, N = 366.4965.781. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm

WebP Image Encode

This is a test of Google's libwebp with the cwebp image encode utility and using a sample 6000x4000 pixel JPEG image as the input. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMP/s, More Is BetterWebP Image Encode 1.2.4Encode Settings: Quality 100, Lossless, Highest Compressionmitigations=offLinux 6.1 Defaults0.21150.4230.63450.8461.0575SE +/- 0.00, N = 3SE +/- 0.00, N = 30.940.931. (CC) gcc options: -fvisibility=hidden -O2 -lm

Apache HBase

This is a benchmark of the Apache HBase non-relational distributed database system inspired from Google's Bigtable. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgRows Per Second, More Is BetterApache HBase 2.5.0Rows: 1000000 - Test: Async Random Write - Clients: 16mitigations=offLinux 6.1 Defaults20K40K60K80K100KSE +/- 824.47, N = 3SE +/- 1004.83, N = 37962978783

Neural Magic DeepSparse

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.1Model: NLP Document Classification, oBERT base uncased on IMDB - Scenario: Asynchronous Multi-Streammitigations=offLinux 6.1 Defaults3691215SE +/- 0.02, N = 3SE +/- 0.05, N = 312.6712.54

ClickHouse

ClickHouse is an open-source, high performance OLAP data management system. This test profile uses ClickHouse's standard benchmark recommendations per https://clickhouse.com/docs/en/operations/performance-test/ with the 100 million rows web analytics dataset. The reported value is the query processing time using the geometric mean of all queries performed. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgQueries Per Minute, Geo Mean, More Is BetterClickHouse 22.5.4.19100M Rows Web Analytics Dataset, Second Runmitigations=offLinux 6.1 Defaults70140210280350SE +/- 1.81, N = 15SE +/- 2.35, N = 15306.76303.65MIN: 24.29 / MAX: 30000MIN: 24.32 / MAX: 300001. ClickHouse server version 22.5.4.19 (official build).

Apache HBase

This is a benchmark of the Apache HBase non-relational distributed database system inspired from Google's Bigtable. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMicroseconds - Average Latency, Fewer Is BetterApache HBase 2.5.0Rows: 1000000 - Test: Async Random Write - Clients: 16mitigations=offLinux 6.1 Defaults4080120160200SE +/- 2.08, N = 3SE +/- 2.67, N = 3200202

Stress-NG

Stress-NG is a Linux stress tool developed by Colin King of Canonical. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: Glibc C String Functionsmitigations=offLinux 6.1 Defaults800K1600K2400K3200K4000KSE +/- 29422.94, N = 3SE +/- 44204.27, N = 33778821.913741573.781. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lrt -lsctp -lz -pthread

OpenVINO

This is a test of the Intel OpenVINO, a toolkit around neural networks, using its built-in benchmarking support and analyzing the throughput and latency for various models. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2022.2.devModel: Person Vehicle Bike Detection FP16 - Device: CPUmitigations=offLinux 6.1 Defaults160320480640800SE +/- 0.34, N = 3SE +/- 3.32, N = 3748.06755.361. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -flto -shared

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2022.2.devModel: Person Vehicle Bike Detection FP16 - Device: CPUmitigations=offLinux 6.1 Defaults3691215SE +/- 0.01, N = 3SE +/- 0.05, N = 310.6810.58MIN: 7.52 / MAX: 54.71MIN: 7.56 / MAX: 57.781. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -flto -shared

Apache Spark

This is a benchmark of Apache Spark with its PySpark interface. Apache Spark is an open-source unified analytics engine for large-scale data processing and dealing with big data. This test profile benchmars the Apache Spark in a single-system configuration using spark-submit. The test makes use of DIYBigData's pyspark-benchmark (https://github.com/DIYBigData/pyspark-benchmark/) for generating of test data and various Apache Spark operations. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterApache Spark 3.3Row Count: 1000000 - Partitions: 500 - Repartition Test Timemitigations=offLinux 6.1 Defaults0.2430.4860.7290.9721.215SE +/- 0.01, N = 7SE +/- 0.03, N = 31.071.08

OSBench

OSBench is a collection of micro-benchmarks for measuring operating system primitives like time to create threads/processes, launching programs, creating files, and memory allocation. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgus Per Event, Fewer Is BetterOSBenchTest: Create Filesmitigations=offLinux 6.1 Defaults246810SE +/- 0.004001, N = 3SE +/- 0.010422, N = 38.2396988.3126711. (CC) gcc options: -lm

Stress-NG

Stress-NG is a Linux stress tool developed by Colin King of Canonical. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: Matrix Mathmitigations=offLinux 6.1 Defaults20K40K60K80K100KSE +/- 1306.38, N = 3SE +/- 692.52, N = 3101602.63102476.691. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lrt -lsctp -lz -pthread

DaCapo Benchmark

This test runs the DaCapo Benchmarks written in Java and intended to test system/CPU performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgmsec, Fewer Is BetterDaCapo Benchmark 9.12-MR1Java Test: Tradesoapmitigations=offLinux 6.1 Defaults400800120016002000SE +/- 11.96, N = 4SE +/- 10.42, N = 2016561642

Facebook RocksDB

OpenBenchmarking.orgOp/s, More Is BetterFacebook RocksDB 7.5.3Test: Random Fill Syncmitigations=offLinux 6.1 Defaults8K16K24K32K40KSE +/- 248.77, N = 3SE +/- 8.74, N = 338607389271. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread

OpenBenchmarking.orgOp/s, More Is BetterFacebook RocksDB 7.5.3Test: Random Fillmitigations=offLinux 6.1 Defaults300K600K900K1200K1500KSE +/- 5444.71, N = 3SE +/- 6463.89, N = 3136726713561171. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread

srsRAN

srsRAN is an open-source LTE/5G software radio suite created by Software Radio Systems (SRS). The srsRAN radio suite was formerly known as srsLTE and can be used for building your own software-defined radio (SDR) 4G/5G mobile network. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgeNb Mb/s, More Is BettersrsRAN 22.04.1Test: 4G PHY_DL_Test 100 PRB MIMO 256-QAMmitigations=offLinux 6.1 Defaults150300450600750SE +/- 2.62, N = 3SE +/- 2.42, N = 3712.1706.41. (CXX) g++ options: -std=c++14 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -ldl -lpthread -lm

MariaDB

This is a MariaDB MySQL database server benchmark making use of mysqlslap. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgQueries Per Second, More Is BetterMariaDB 10.8.2Clients: 32mitigations=offLinux 6.1 Defaults7001400210028003500SE +/- 9.76, N = 3SE +/- 28.86, N = 3333033041. (CXX) g++ options: -pie -fPIC -fstack-protector -O3 -lnuma -lpcre2-8 -lcrypt -lz -lm -lssl -lcrypto -lpthread -ldl

srsRAN

srsRAN is an open-source LTE/5G software radio suite created by Software Radio Systems (SRS). The srsRAN radio suite was formerly known as srsLTE and can be used for building your own software-defined radio (SDR) 4G/5G mobile network. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgeNb Mb/s, More Is BettersrsRAN 22.04.1Test: 5G PHY_DL_NR Test 52 PRB SISO 64-QAMmitigations=offLinux 6.1 Defaults50100150200250SE +/- 1.89, N = 3SE +/- 1.46, N = 3225.8227.51. (CXX) g++ options: -std=c++14 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -ldl -lpthread -lm

KeyDB

A benchmark of KeyDB as a multi-threaded fork of the Redis server. The KeyDB benchmark is conducted using memtier-benchmark. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgOps/sec, More Is BetterKeyDB 6.2.0mitigations=offLinux 6.1 Defaults300K600K900K1200K1500KSE +/- 3433.04, N = 3SE +/- 3644.79, N = 31518638.231507777.211. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre

Facebook RocksDB

OpenBenchmarking.orgOp/s, More Is BetterFacebook RocksDB 7.5.3Test: Update Randommitigations=offLinux 6.1 Defaults160K320K480K640K800KSE +/- 4236.67, N = 3SE +/- 7415.12, N = 37645587700171. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread

Apache CouchDB

This is a bulk insertion benchmark of Apache CouchDB. CouchDB is a document-oriented NoSQL database implemented in Erlang. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterApache CouchDB 3.2.2Bulk Size: 100 - Inserts: 3000 - Rounds: 30mitigations=offLinux 6.1 Defaults4080120160200SE +/- 0.35, N = 3SE +/- 0.13, N = 3167.47168.661. (CXX) g++ options: -std=c++17 -lmozjs-91 -lm -lei -fPIC -MMD

Renaissance

Renaissance is a suite of benchmarks designed to test the Java JVM from Apache Spark to a Twitter-like service to Scala and other features. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterRenaissance 0.14Test: Savina Reactors.IOmitigations=offLinux 6.1 Defaults9001800270036004500SE +/- 3.06, N = 3SE +/- 45.21, N = 34159.14188.3MIN: 4153.01 / MAX: 5832.37MIN: 4132.82 / MAX: 5963.25

Dragonflydb

Dragonfly is an open-source database server that is a "modern Redis replacement" that aims to be the fastest memory store while being compliant with the Redis and Memcached protocols. For benchmarking Dragonfly, Memtier_benchmark is used as a NoSQL Redis/Memcache traffic generation plus benchmarking tool developed by Redis Labs. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgOps/sec, More Is BetterDragonflydb 0.6Clients: 200 - Set To Get Ratio: 1:1mitigations=offLinux 6.1 Defaults1.4M2.8M4.2M5.6M7MSE +/- 49539.07, N = 3SE +/- 42144.49, N = 36650409.606696607.281. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre

OpenVINO

This is a test of the Intel OpenVINO, a toolkit around neural networks, using its built-in benchmarking support and analyzing the throughput and latency for various models. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2022.2.devModel: Weld Porosity Detection FP16-INT8 - Device: CPUmitigations=offLinux 6.1 Defaults48121620SE +/- 0.03, N = 3SE +/- 0.02, N = 314.5114.41MIN: 7.71 / MAX: 127.13MIN: 7.04 / MAX: 135.231. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -flto -shared

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2022.2.devModel: Weld Porosity Detection FP16-INT8 - Device: CPUmitigations=offLinux 6.1 Defaults400800120016002000SE +/- 3.13, N = 3SE +/- 1.49, N = 31653.361664.641. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -flto -shared

Stress-NG

Stress-NG is a Linux stress tool developed by Colin King of Canonical. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: IO_uringmitigations=offLinux 6.1 Defaults6K12K18K24K30KSE +/- 16.82, N = 3SE +/- 6.02, N = 327589.4327773.351. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lrt -lsctp -lz -pthread

ASTC Encoder

ASTC Encoder (astcenc) is for the Adaptive Scalable Texture Compression (ASTC) format commonly used with OpenGL, OpenGL ES, and Vulkan graphics APIs. This test profile does a coding test of both compression/decompression. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMT/s, More Is BetterASTC Encoder 4.0Preset: Exhaustivemitigations=offLinux 6.1 Defaults0.32880.65760.98641.31521.644SE +/- 0.0029, N = 3SE +/- 0.0062, N = 31.46121.45161. (CXX) g++ options: -O3 -flto -pthread

srsRAN

srsRAN is an open-source LTE/5G software radio suite created by Software Radio Systems (SRS). The srsRAN radio suite was formerly known as srsLTE and can be used for building your own software-defined radio (SDR) 4G/5G mobile network. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgeNb Mb/s, More Is BettersrsRAN 22.04.1Test: 4G PHY_DL_Test 100 PRB MIMO 64-QAMmitigations=offLinux 6.1 Defaults140280420560700SE +/- 4.97, N = 3SE +/- 5.43, N = 3639.1643.21. (CXX) g++ options: -std=c++14 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -ldl -lpthread -lm

PostgreSQL

This is a benchmark of PostgreSQL using the integrated pgbench for facilitating the database benchmarks. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL 15Scaling Factor: 100 - Clients: 50 - Mode: Read Onlymitigations=offLinux 6.1 Defaults300K600K900K1200K1500KSE +/- 6479.50, N = 3SE +/- 10486.93, N = 3126865712608851. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL 15Scaling Factor: 100 - Clients: 100 - Mode: Read Onlymitigations=offLinux 6.1 Defaults300K600K900K1200K1500KSE +/- 13433.82, N = 4SE +/- 3039.24, N = 3123826512306881. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm

InfluxDB

This is a benchmark of the InfluxDB open-source time-series database optimized for fast, high-availability storage for IoT and other use-cases. The InfluxDB test profile makes use of InfluxDB Inch for facilitating the benchmarks. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgval/sec, More Is BetterInfluxDB 1.8.2Concurrent Streams: 4 - Batch Size: 10000 - Tags: 2,5000,1 - Points Per Series: 10000mitigations=offLinux 6.1 Defaults600K1200K1800K2400K3000KSE +/- 26949.91, N = 6SE +/- 11434.48, N = 32680364.82696816.3

srsRAN

srsRAN is an open-source LTE/5G software radio suite created by Software Radio Systems (SRS). The srsRAN radio suite was formerly known as srsLTE and can be used for building your own software-defined radio (SDR) 4G/5G mobile network. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgUE Mb/s, More Is BettersrsRAN 22.04.1Test: 4G PHY_DL_Test 100 PRB MIMO 64-QAMmitigations=offLinux 6.1 Defaults4080120160200SE +/- 1.23, N = 3SE +/- 0.17, N = 3180.4181.51. (CXX) g++ options: -std=c++14 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -ldl -lpthread -lm

Stress-NG

Stress-NG is a Linux stress tool developed by Colin King of Canonical. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: Glibc Qsort Data Sortingmitigations=offLinux 6.1 Defaults90180270360450SE +/- 1.72, N = 3SE +/- 2.44, N = 3411.91414.371. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lrt -lsctp -lz -pthread

Neural Magic DeepSparse

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.1Model: CV Detection,YOLOv5s COCO - Scenario: Asynchronous Multi-Streammitigations=offLinux 6.1 Defaults20406080100SE +/- 0.47, N = 3SE +/- 0.27, N = 375.2075.64

Apache Spark

This is a benchmark of Apache Spark with its PySpark interface. Apache Spark is an open-source unified analytics engine for large-scale data processing and dealing with big data. This test profile benchmars the Apache Spark in a single-system configuration using spark-submit. The test makes use of DIYBigData's pyspark-benchmark (https://github.com/DIYBigData/pyspark-benchmark/) for generating of test data and various Apache Spark operations. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterApache Spark 3.3Row Count: 1000000 - Partitions: 500 - Calculate Pi Benchmark Using Dataframemitigations=offLinux 6.1 Defaults0.7651.532.2953.063.825SE +/- 0.06, N = 7SE +/- 0.10, N = 33.403.38

DaCapo Benchmark

This test runs the DaCapo Benchmarks written in Java and intended to test system/CPU performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgmsec, Fewer Is BetterDaCapo Benchmark 9.12-MR1Java Test: Tradebeansmitigations=offLinux 6.1 Defaults400800120016002000SE +/- 10.14, N = 4SE +/- 13.02, N = 417191709

Neural Magic DeepSparse

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.1Model: NLP Text Classification, DistilBERT mnli - Scenario: Asynchronous Multi-Streammitigations=offLinux 6.1 Defaults306090120150SE +/- 1.06, N = 3SE +/- 1.09, N = 3113.43112.77

Bork File Encrypter

Bork is a small, cross-platform file encryption utility. It is written in Java and designed to be included along with the files it encrypts for long-term storage. This test measures the amount of time it takes to encrypt a sample file. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterBork File Encrypter 1.4File Encryption Timemitigations=offLinux 6.1 Defaults1.09012.18023.27034.36045.4505SE +/- 0.030, N = 3SE +/- 0.033, N = 34.8174.845

Renaissance

Renaissance is a suite of benchmarks designed to test the Java JVM from Apache Spark to a Twitter-like service to Scala and other features. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterRenaissance 0.14Test: Apache Spark Bayesmitigations=offLinux 6.1 Defaults150300450600750SE +/- 3.25, N = 3SE +/- 1.25, N = 3697.8701.8MIN: 502.51 / MAX: 703.67MIN: 510.43 / MAX: 704.23

OpenVINO

This is a test of the Intel OpenVINO, a toolkit around neural networks, using its built-in benchmarking support and analyzing the throughput and latency for various models. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2022.2.devModel: Person Detection FP32 - Device: CPUmitigations=offLinux 6.1 Defaults5001000150020002500SE +/- 6.69, N = 3SE +/- 4.82, N = 32236.032223.37MIN: 1702.71 / MAX: 3129.63MIN: 1691.82 / MAX: 2995.511. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -flto -shared

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2022.2.devModel: Person Detection FP32 - Device: CPUmitigations=offLinux 6.1 Defaults0.80551.6112.41653.2224.0275SE +/- 0.01, N = 3SE +/- 0.00, N = 33.563.581. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -flto -shared

Neural Magic DeepSparse

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.1Model: CV Detection,YOLOv5s COCO - Scenario: Asynchronous Multi-Streammitigations=offLinux 6.1 Defaults4080120160200SE +/- 0.78, N = 3SE +/- 0.69, N = 3158.91158.06

OpenVINO

This is a test of the Intel OpenVINO, a toolkit around neural networks, using its built-in benchmarking support and analyzing the throughput and latency for various models. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2022.2.devModel: Vehicle Detection FP16 - Device: CPUmitigations=offLinux 6.1 Defaults80160240320400SE +/- 0.52, N = 3SE +/- 0.29, N = 3370.54372.541. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -flto -shared

Neural Magic DeepSparse

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.1Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Asynchronous Multi-Streammitigations=offLinux 6.1 Defaults2004006008001000SE +/- 1.94, N = 3SE +/- 3.70, N = 3952.75947.67

Facebook RocksDB

OpenBenchmarking.orgOp/s, More Is BetterFacebook RocksDB 7.5.3Test: Sequential Fillmitigations=offLinux 6.1 Defaults300K600K900K1200K1500KSE +/- 2086.01, N = 3SE +/- 17853.48, N = 3149461015024311. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread

OpenVINO

This is a test of the Intel OpenVINO, a toolkit around neural networks, using its built-in benchmarking support and analyzing the throughput and latency for various models. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2022.2.devModel: Vehicle Detection FP16 - Device: CPUmitigations=offLinux 6.1 Defaults510152025SE +/- 0.03, N = 3SE +/- 0.02, N = 321.5621.45MIN: 11.75 / MAX: 89.39MIN: 11.89 / MAX: 881. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -flto -shared

Neural Magic DeepSparse

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.1Model: NLP Text Classification, DistilBERT mnli - Scenario: Asynchronous Multi-Streammitigations=offLinux 6.1 Defaults20406080100SE +/- 1.02, N = 3SE +/- 1.00, N = 3105.73106.27

SVT-AV1

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 1.2Encoder Mode: Preset 4 - Input: Bosphorus 4Kmitigations=offLinux 6.1 Defaults0.68831.37662.06492.75323.4415SE +/- 0.005, N = 3SE +/- 0.016, N = 33.0443.0591. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq

OpenVINO

This is a test of the Intel OpenVINO, a toolkit around neural networks, using its built-in benchmarking support and analyzing the throughput and latency for various models. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2022.2.devModel: Vehicle Detection FP16-INT8 - Device: CPUmitigations=offLinux 6.1 Defaults2004006008001000SE +/- 1.18, N = 3SE +/- 0.49, N = 3914.62919.041. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -flto -shared

Neural Magic DeepSparse

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.1Model: NLP Token Classification, BERT base uncased conll2003 - Scenario: Asynchronous Multi-Streammitigations=offLinux 6.1 Defaults3691215SE +/- 0.04, N = 3SE +/- 0.05, N = 312.4512.51

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.1Model: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Scenario: Asynchronous Multi-Streammitigations=offLinux 6.1 Defaults1122334455SE +/- 0.12, N = 3SE +/- 0.16, N = 350.3050.53

OpenVINO

This is a test of the Intel OpenVINO, a toolkit around neural networks, using its built-in benchmarking support and analyzing the throughput and latency for various models. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2022.2.devModel: Vehicle Detection FP16-INT8 - Device: CPUmitigations=offLinux 6.1 Defaults246810SE +/- 0.01, N = 3SE +/- 0.00, N = 38.748.70MIN: 5.93 / MAX: 61MIN: 5.93 / MAX: 62.711. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -flto -shared

Facebook RocksDB

OpenBenchmarking.orgOp/s, More Is BetterFacebook RocksDB 7.5.3Test: Read Random Write Randommitigations=offLinux 6.1 Defaults800K1600K2400K3200K4000KSE +/- 15106.42, N = 3SE +/- 3258.62, N = 3362464436411551. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread

OSBench

OSBench is a collection of micro-benchmarks for measuring operating system primitives like time to create threads/processes, launching programs, creating files, and memory allocation. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgus Per Event, Fewer Is BetterOSBenchTest: Launch Programsmitigations=offLinux 6.1 Defaults612182430SE +/- 0.16, N = 3SE +/- 0.07, N = 323.8923.991. (CC) gcc options: -lm

libavif avifenc

This is a test of the AOMedia libavif library testing the encoding of a JPEG image to AV1 Image Format (AVIF). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.11Encoder Speed: 0mitigations=offLinux 6.1 Defaults1632486480SE +/- 0.18, N = 3SE +/- 0.12, N = 370.0470.341. (CXX) g++ options: -O3 -fPIC -lm

OpenBenchmarking.orgSeconds, Fewer Is Betterlibavif avifenc 0.11Encoder Speed: 2mitigations=offLinux 6.1 Defaults816243240SE +/- 0.09, N = 3SE +/- 0.28, N = 334.1134.261. (CXX) g++ options: -O3 -fPIC -lm

SVT-AV1

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 1.2Encoder Mode: Preset 12 - Input: Bosphorus 4Kmitigations=offLinux 6.1 Defaults50100150200250SE +/- 0.64, N = 3SE +/- 1.49, N = 3217.73216.781. (CXX) g++ options: -march=native -mno-avx -mavx2 -mavx512f -mavx512bw -mavx512dq

Stargate Digital Audio Workstation

Stargate is an open-source, cross-platform digital audio workstation (DAW) software package with "a unique and carefully curated experience" with scalability from old systems up through modern multi-core systems. Stargate is GPLv3 licensed and makes use of Qt5 (PyQt5) for its user-interface. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgRender Ratio, More Is BetterStargate Digital Audio Workstation 22.11.5Sample Rate: 192000 - Buffer Size: 1024mitigations=offLinux 6.1 Defaults0.68551.3712.05652.7423.4275SE +/- 0.003982, N = 3SE +/- 0.002663, N = 33.0467163.0335331. (CXX) g++ options: -lpthread -lsndfile -lm -O3 -march=native -ffast-math -funroll-loops -fstrength-reduce -fstrict-aliasing -finline-functions

Cryptsetup

This is a test profile for running the cryptsetup benchmark to report on the system's cryptography performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/s, More Is BetterCryptsetupAES-XTS 256b Encryptionmitigations=offLinux 6.1 Defaults15003000450060007500SE +/- 24.85, N = 3SE +/- 15.48, N = 37078.47107.7

srsRAN

srsRAN is an open-source LTE/5G software radio suite created by Software Radio Systems (SRS). The srsRAN radio suite was formerly known as srsLTE and can be used for building your own software-defined radio (SDR) 4G/5G mobile network. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSamples / Second, More Is BettersrsRAN 22.04.1Test: OFDM_Testmitigations=offLinux 6.1 Defaults40M80M120M160M200MSE +/- 642910.05, N = 3SE +/- 866666.67, N = 32006000001998333331. (CXX) g++ options: -std=c++14 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -ldl -lpthread -lm

Aircrack-ng

Aircrack-ng is a tool for assessing WiFi/WLAN network security. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgk/s, More Is BetterAircrack-ng 1.7mitigations=offLinux 6.1 Defaults15K30K45K60K75KSE +/- 47.65, N = 3SE +/- 164.84, N = 369948.1969682.141. (CXX) g++ options: -std=gnu++17 -O3 -fvisibility=hidden -fcommon -rdynamic -lnl-3 -lnl-genl-3 -lpcre -lsqlite3 -lpthread -lz -lssl -lcrypto -lhwloc -ldl -lm -pthread

Stress-NG

Stress-NG is a Linux stress tool developed by Colin King of Canonical. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: Memory Copyingmitigations=offLinux 6.1 Defaults16003200480064008000SE +/- 24.63, N = 3SE +/- 42.34, N = 37275.147302.741. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lrt -lsctp -lz -pthread

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: Cryptomitigations=offLinux 6.1 Defaults9K18K27K36K45KSE +/- 226.79, N = 3SE +/- 251.10, N = 342952.3242801.841. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lrt -lsctp -lz -pthread

Cryptsetup

This is a test profile for running the cryptsetup benchmark to report on the system's cryptography performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/s, More Is BetterCryptsetupAES-XTS 256b Decryptionmitigations=offLinux 6.1 Defaults15003000450060007500SE +/- 24.01, N = 3SE +/- 15.14, N = 37083.97108.8

DaCapo Benchmark

This test runs the DaCapo Benchmarks written in Java and intended to test system/CPU performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgmsec, Fewer Is BetterDaCapo Benchmark 9.12-MR1Java Test: Jythonmitigations=offLinux 6.1 Defaults400800120016002000SE +/- 13.09, N = 10SE +/- 11.93, N = 417151709

Dragonflydb

Dragonfly is an open-source database server that is a "modern Redis replacement" that aims to be the fastest memory store while being compliant with the Redis and Memcached protocols. For benchmarking Dragonfly, Memtier_benchmark is used as a NoSQL Redis/Memcache traffic generation plus benchmarking tool developed by Redis Labs. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgOps/sec, More Is BetterDragonflydb 0.6Clients: 50 - Set To Get Ratio: 1:5mitigations=offLinux 6.1 Defaults1.4M2.8M4.2M5.6M7MSE +/- 31744.79, N = 3SE +/- 16253.11, N = 36765462.426742394.011. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre

AOM AV1

This is a test of the AOMedia AV1 encoder (libaom) developed by AOMedia and Google as the AV1 Codec Library. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.5Encoder Mode: Speed 4 Two-Pass - Input: Bosphorus 4Kmitigations=offLinux 6.1 Defaults3691215SE +/- 0.04, N = 3SE +/- 0.01, N = 312.0512.091. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm

Renaissance

Renaissance is a suite of benchmarks designed to test the Java JVM from Apache Spark to a Twitter-like service to Scala and other features. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterRenaissance 0.14Test: Random Forestmitigations=offLinux 6.1 Defaults80160240320400SE +/- 1.71, N = 3SE +/- 1.64, N = 3378.8380.0MIN: 350.48 / MAX: 457.26MIN: 353.4 / MAX: 457.98

ClickHouse

ClickHouse is an open-source, high performance OLAP data management system. This test profile uses ClickHouse's standard benchmark recommendations per https://clickhouse.com/docs/en/operations/performance-test/ with the 100 million rows web analytics dataset. The reported value is the query processing time using the geometric mean of all queries performed. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgQueries Per Minute, Geo Mean, More Is BetterClickHouse 22.5.4.19100M Rows Web Analytics Dataset, First Run / Cold Cachemitigations=offLinux 6.1 Defaults70140210280350SE +/- 2.21, N = 15SE +/- 1.97, N = 15299.11300.04MIN: 23.78 / MAX: 30000MIN: 23.58 / MAX: 300001. ClickHouse server version 22.5.4.19 (official build).

WebP Image Encode

This is a test of Google's libwebp with the cwebp image encode utility and using a sample 6000x4000 pixel JPEG image as the input. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMP/s, More Is BetterWebP Image Encode 1.2.4Encode Settings: Quality 100mitigations=offLinux 6.1 Defaults48121620SE +/- 0.01, N = 3SE +/- 0.01, N = 316.6416.691. (CC) gcc options: -fvisibility=hidden -O2 -lm

Neural Magic DeepSparse

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.1Model: NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Scenario: Asynchronous Multi-Streammitigations=offLinux 6.1 Defaults50100150200250SE +/- 0.27, N = 3SE +/- 0.44, N = 3237.44236.76

OpenVINO

This is a test of the Intel OpenVINO, a toolkit around neural networks, using its built-in benchmarking support and analyzing the throughput and latency for various models. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2022.2.devModel: Person Detection FP16 - Device: CPUmitigations=offLinux 6.1 Defaults0.811.622.433.244.05SE +/- 0.00, N = 3SE +/- 0.00, N = 33.603.591. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -flto -shared

Apache Spark

This is a benchmark of Apache Spark with its PySpark interface. Apache Spark is an open-source unified analytics engine for large-scale data processing and dealing with big data. This test profile benchmars the Apache Spark in a single-system configuration using spark-submit. The test makes use of DIYBigData's pyspark-benchmark (https://github.com/DIYBigData/pyspark-benchmark/) for generating of test data and various Apache Spark operations. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterApache Spark 3.3Row Count: 1000000 - Partitions: 500 - Calculate Pi Benchmarkmitigations=offLinux 6.1 Defaults1224364860SE +/- 0.11, N = 7SE +/- 0.24, N = 351.4251.29

SQLite Speedtest

This is a benchmark of SQLite's speedtest1 benchmark program with an increased problem size of 1,000. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterSQLite Speedtest 3.30Timed Time - Size 1,000mitigations=offLinux 6.1 Defaults714212835SE +/- 0.01, N = 3SE +/- 0.04, N = 331.3931.311. (CC) gcc options: -O2 -lz

nginx

This is a benchmark of the lightweight Nginx HTTP(S) web-server. This Nginx web server benchmark test profile makes use of the wrk program for facilitating the HTTP requests over a fixed period time with a configurable number of concurrent clients/connections. HTTPS with a self-signed OpenSSL certificate is used by this test for local benchmarking. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgRequests Per Second, More Is Betternginx 1.23.2Connections: 100mitigations=offLinux 6.1 Defaults40K80K120K160K200KSE +/- 1384.83, N = 3SE +/- 1157.09, N = 3207784.83207280.771. (CC) gcc options: -lluajit-5.1 -lm -lssl -lcrypto -lpthread -ldl -std=c99 -O2

Stargate Digital Audio Workstation

Stargate is an open-source, cross-platform digital audio workstation (DAW) software package with "a unique and carefully curated experience" with scalability from old systems up through modern multi-core systems. Stargate is GPLv3 licensed and makes use of Qt5 (PyQt5) for its user-interface. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgRender Ratio, More Is BetterStargate Digital Audio Workstation 22.11.5Sample Rate: 192000 - Buffer Size: 512mitigations=offLinux 6.1 Defaults0.5931.1861.7792.3722.965SE +/- 0.006256, N = 3SE +/- 0.005802, N = 32.6292252.6355901. (CXX) g++ options: -lpthread -lsndfile -lm -O3 -march=native -ffast-math -funroll-loops -fstrength-reduce -fstrict-aliasing -finline-functions

Cryptsetup

This is a test profile for running the cryptsetup benchmark to report on the system's cryptography performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/s, More Is BetterCryptsetupAES-XTS 512b Encryptionmitigations=offLinux 6.1 Defaults14002800420056007000SE +/- 22.60, N = 3SE +/- 16.24, N = 36508.06523.5

WebP Image Encode

This is a test of Google's libwebp with the cwebp image encode utility and using a sample 6000x4000 pixel JPEG image as the input. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMP/s, More Is BetterWebP Image Encode 1.2.4Encode Settings: Defaultmitigations=offLinux 6.1 Defaults612182430SE +/- 0.02, N = 3SE +/- 0.01, N = 325.8225.881. (CC) gcc options: -fvisibility=hidden -O2 -lm

Java Gradle Build

This test runs Java software project builds using the Gradle build system. It is intended to give developers an idea as to the build performance for development activities and build servers. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterJava Gradle BuildGradle Build: Reactormitigations=offLinux 6.1 Defaults306090120150SE +/- 1.41, N = 12SE +/- 1.24, N = 3141.42141.72

OpenVINO

This is a test of the Intel OpenVINO, a toolkit around neural networks, using its built-in benchmarking support and analyzing the throughput and latency for various models. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2022.2.devModel: Weld Porosity Detection FP16 - Device: CPUmitigations=offLinux 6.1 Defaults100200300400500SE +/- 0.03, N = 3SE +/- 0.35, N = 3472.81473.811. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -flto -shared

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2022.2.devModel: Weld Porosity Detection FP16 - Device: CPUmitigations=offLinux 6.1 Defaults1122334455SE +/- 0.01, N = 3SE +/- 0.04, N = 350.6750.57MIN: 30.2 / MAX: 197.71MIN: 22.73 / MAX: 185.961. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -flto -shared

nginx

This is a benchmark of the lightweight Nginx HTTP(S) web-server. This Nginx web server benchmark test profile makes use of the wrk program for facilitating the HTTP requests over a fixed period time with a configurable number of concurrent clients/connections. HTTPS with a self-signed OpenSSL certificate is used by this test for local benchmarking. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgRequests Per Second, More Is Betternginx 1.23.2Connections: 500mitigations=offLinux 6.1 Defaults40K80K120K160K200KSE +/- 812.38, N = 3SE +/- 845.21, N = 3202777.49203178.401. (CC) gcc options: -lluajit-5.1 -lm -lssl -lcrypto -lpthread -ldl -std=c99 -O2

Cryptsetup

This is a test profile for running the cryptsetup benchmark to report on the system's cryptography performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/s, More Is BetterCryptsetupAES-XTS 512b Decryptionmitigations=offLinux 6.1 Defaults14002800420056007000SE +/- 21.42, N = 3SE +/- 16.52, N = 36490.66503.1

nginx

This is a benchmark of the lightweight Nginx HTTP(S) web-server. This Nginx web server benchmark test profile makes use of the wrk program for facilitating the HTTP requests over a fixed period time with a configurable number of concurrent clients/connections. HTTPS with a self-signed OpenSSL certificate is used by this test for local benchmarking. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgRequests Per Second, More Is Betternginx 1.23.2Connections: 1000mitigations=offLinux 6.1 Defaults40K80K120K160K200KSE +/- 586.42, N = 3SE +/- 221.20, N = 3187868.73187512.541. (CC) gcc options: -lluajit-5.1 -lm -lssl -lcrypto -lpthread -ldl -std=c99 -O2

Cryptsetup

This is a test profile for running the cryptsetup benchmark to report on the system's cryptography performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgIterations Per Second, More Is BetterCryptsetupPBKDF2-whirlpoolmitigations=offLinux 6.1 Defaults300K600K900K1200K1500KSE +/- 568.00, N = 3SE +/- 1131.33, N = 313369011334634

srsRAN

srsRAN is an open-source LTE/5G software radio suite created by Software Radio Systems (SRS). The srsRAN radio suite was formerly known as srsLTE and can be used for building your own software-defined radio (SDR) 4G/5G mobile network. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgUE Mb/s, More Is BettersrsRAN 22.04.1Test: 4G PHY_DL_Test 100 PRB SISO 256-QAMmitigations=offLinux 6.1 Defaults50100150200250SE +/- 0.50, N = 3SE +/- 1.56, N = 3245.5245.11. (CXX) g++ options: -std=c++14 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -ldl -lpthread -lm

OpenVINO

This is a test of the Intel OpenVINO, a toolkit around neural networks, using its built-in benchmarking support and analyzing the throughput and latency for various models. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2022.2.devModel: Face Detection FP16-INT8 - Device: CPUmitigations=offLinux 6.1 Defaults90180270360450SE +/- 0.58, N = 3SE +/- 0.12, N = 3438.11437.50MIN: 270.11 / MAX: 1300.72MIN: 270.24 / MAX: 992.111. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -flto -shared

InfluxDB

This is a benchmark of the InfluxDB open-source time-series database optimized for fast, high-availability storage for IoT and other use-cases. The InfluxDB test profile makes use of InfluxDB Inch for facilitating the benchmarks. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgval/sec, More Is BetterInfluxDB 1.8.2Concurrent Streams: 64 - Batch Size: 10000 - Tags: 2,5000,1 - Points Per Series: 10000mitigations=offLinux 6.1 Defaults600K1200K1800K2400K3000KSE +/- 3790.37, N = 3SE +/- 5792.62, N = 32976169.82980179.4

Timed Gem5 Compilation

This test times how long it takes to compile Gem5. Gem5 is a simulator for computer system architecture research. Gem5 is widely used for computer architecture research within the industry, academia, and more. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Gem5 Compilation 21.2Time To Compilemitigations=offLinux 6.1 Defaults4080120160200SE +/- 0.76, N = 3SE +/- 0.24, N = 3174.41174.64

Cryptsetup

This is a test profile for running the cryptsetup benchmark to report on the system's cryptography performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/s, More Is BetterCryptsetupSerpent-XTS 512b Encryptionmitigations=offLinux 6.1 Defaults2004006008001000SE +/- 0.80, N = 2SE +/- 0.62, N = 31010.41011.7

OpenVINO

This is a test of the Intel OpenVINO, a toolkit around neural networks, using its built-in benchmarking support and analyzing the throughput and latency for various models. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2022.2.devModel: Machine Translation EN To DE FP16 - Device: CPUmitigations=offLinux 6.1 Defaults1428425670SE +/- 0.10, N = 3SE +/- 0.06, N = 363.8763.951. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -flto -shared

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2022.2.devModel: Person Detection FP16 - Device: CPUmitigations=offLinux 6.1 Defaults5001000150020002500SE +/- 2.24, N = 3SE +/- 1.23, N = 32209.762212.51MIN: 1693.41 / MAX: 2903.99MIN: 1666.75 / MAX: 2938.51. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -flto -shared

Stress-NG

Stress-NG is a Linux stress tool developed by Colin King of Canonical. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: Vector Mathmitigations=offLinux 6.1 Defaults30K60K90K120K150KSE +/- 179.45, N = 3SE +/- 174.14, N = 3134263.01134428.581. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lrt -lsctp -lz -pthread

nginx

This is a benchmark of the lightweight Nginx HTTP(S) web-server. This Nginx web server benchmark test profile makes use of the wrk program for facilitating the HTTP requests over a fixed period time with a configurable number of concurrent clients/connections. HTTPS with a self-signed OpenSSL certificate is used by this test for local benchmarking. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgRequests Per Second, More Is Betternginx 1.23.2Connections: 200mitigations=offLinux 6.1 Defaults50K100K150K200K250KSE +/- 702.24, N = 3SE +/- 806.08, N = 3211212.37211451.291. (CC) gcc options: -lluajit-5.1 -lm -lssl -lcrypto -lpthread -ldl -std=c99 -O2

OpenVINO

This is a test of the Intel OpenVINO, a toolkit around neural networks, using its built-in benchmarking support and analyzing the throughput and latency for various models. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2022.2.devModel: Face Detection FP16-INT8 - Device: CPUmitigations=offLinux 6.1 Defaults48121620SE +/- 0.03, N = 3SE +/- 0.00, N = 318.2418.261. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -flto -shared

OSBench

OSBench is a collection of micro-benchmarks for measuring operating system primitives like time to create threads/processes, launching programs, creating files, and memory allocation. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgus Per Event, Fewer Is BetterOSBenchTest: Create Threadsmitigations=offLinux 6.1 Defaults3691215SE +/- 0.101921, N = 3SE +/- 0.059195, N = 39.1067959.1163321. (CC) gcc options: -lm

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 5.18Build: defconfigmitigations=offLinux 6.1 Defaults816243240SE +/- 0.43, N = 3SE +/- 0.26, N = 335.6935.65

Stress-NG

Stress-NG is a Linux stress tool developed by Colin King of Canonical. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: CPU Stressmitigations=offLinux 6.1 Defaults11K22K33K44K55KSE +/- 646.04, N = 3SE +/- 433.83, N = 351447.4851499.151. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lrt -lsctp -lz -pthread

ASTC Encoder

ASTC Encoder (astcenc) is for the Adaptive Scalable Texture Compression (ASTC) format commonly used with OpenGL, OpenGL ES, and Vulkan graphics APIs. This test profile does a coding test of both compression/decompression. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMT/s, More Is BetterASTC Encoder 4.0Preset: Fastmitigations=offLinux 6.1 Defaults70140210280350SE +/- 0.26, N = 3SE +/- 0.16, N = 3344.76345.111. (CXX) g++ options: -O3 -flto -pthread

srsRAN

srsRAN is an open-source LTE/5G software radio suite created by Software Radio Systems (SRS). The srsRAN radio suite was formerly known as srsLTE and can be used for building your own software-defined radio (SDR) 4G/5G mobile network. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgUE Mb/s, More Is BettersrsRAN 22.04.1Test: 4G PHY_DL_Test 100 PRB MIMO 256-QAMmitigations=offLinux 6.1 Defaults4080120160200SE +/- 0.32, N = 3SE +/- 0.12, N = 3201.8202.01. (CXX) g++ options: -std=c++14 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -ldl -lpthread -lm

AOM AV1

This is a test of the AOMedia AV1 encoder (libaom) developed by AOMedia and Google as the AV1 Codec Library. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 3.5Encoder Mode: Speed 6 Two-Pass - Input: Bosphorus 4Kmitigations=offLinux 6.1 Defaults510152025SE +/- 0.07, N = 3SE +/- 0.06, N = 321.6821.701. (CXX) g++ options: -O3 -std=c++11 -U_FORTIFY_SOURCE -lm

BRL-CAD

BRL-CAD is a cross-platform, open-source solid modeling system with built-in benchmark mode. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgVGR Performance Metric, More Is BetterBRL-CAD 7.32.6VGR Performance Metricmitigations=offLinux 6.1 Defaults100K200K300K400K500K4633654629451. (CXX) g++ options: -std=c++11 -pipe -fvisibility=hidden -fno-strict-aliasing -fno-common -fexceptions -ftemplate-depth-128 -m64 -ggdb3 -O3 -fipa-pta -fstrength-reduce -finline-functions -flto -pedantic -ldl -lm

ASTC Encoder

ASTC Encoder (astcenc) is for the Adaptive Scalable Texture Compression (ASTC) format commonly used with OpenGL, OpenGL ES, and Vulkan graphics APIs. This test profile does a coding test of both compression/decompression. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMT/s, More Is BetterASTC Encoder 4.0Preset: Mediummitigations=offLinux 6.1 Defaults306090120150SE +/- 0.12, N = 3SE +/- 0.19, N = 3126.26126.151. (CXX) g++ options: -O3 -flto -pthread

Apache CouchDB

This is a bulk insertion benchmark of Apache CouchDB. CouchDB is a document-oriented NoSQL database implemented in Erlang. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterApache CouchDB 3.2.2Bulk Size: 300 - Inserts: 3000 - Rounds: 30mitigations=offLinux 6.1 Defaults60120180240300SE +/- 0.88, N = 3SE +/- 2.18, N = 3287.31287.571. (CXX) g++ options: -std=c++17 -lmozjs-91 -lm -lei -fPIC -MMD

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 5.18Build: allmodconfigmitigations=offLinux 6.1 Defaults90180270360450SE +/- 0.35, N = 3SE +/- 0.44, N = 3410.34410.70

Cryptsetup

This is a test profile for running the cryptsetup benchmark to report on the system's cryptography performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/s, More Is BetterCryptsetupSerpent-XTS 256b Decryptionmitigations=offLinux 6.1 Defaults2004006008001000SE +/- 0.95, N = 3SE +/- 0.74, N = 31042.71043.6

OpenVINO

This is a test of the Intel OpenVINO, a toolkit around neural networks, using its built-in benchmarking support and analyzing the throughput and latency for various models. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2022.2.devModel: Machine Translation EN To DE FP16 - Device: CPUmitigations=offLinux 6.1 Defaults306090120150SE +/- 0.22, N = 3SE +/- 0.10, N = 3125.10125.00MIN: 90.23 / MAX: 327.3MIN: 88.24 / MAX: 330.61. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -flto -shared

Apache Spark

This is a benchmark of Apache Spark with its PySpark interface. Apache Spark is an open-source unified analytics engine for large-scale data processing and dealing with big data. This test profile benchmars the Apache Spark in a single-system configuration using spark-submit. The test makes use of DIYBigData's pyspark-benchmark (https://github.com/DIYBigData/pyspark-benchmark/) for generating of test data and various Apache Spark operations. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterApache Spark 3.3Row Count: 1000000 - Partitions: 100 - Calculate Pi Benchmarkmitigations=offLinux 6.1 Defaults1224364860SE +/- 0.08, N = 15SE +/- 0.07, N = 1551.2551.21

Dragonflydb

Dragonfly is an open-source database server that is a "modern Redis replacement" that aims to be the fastest memory store while being compliant with the Redis and Memcached protocols. For benchmarking Dragonfly, Memtier_benchmark is used as a NoSQL Redis/Memcache traffic generation plus benchmarking tool developed by Redis Labs. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgOps/sec, More Is BetterDragonflydb 0.6Clients: 200 - Set To Get Ratio: 5:1mitigations=offLinux 6.1 Defaults1.4M2.8M4.2M5.6M7MSE +/- 46202.73, N = 3SE +/- 19398.36, N = 36436491.376440950.321. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre

OpenVINO

This is a test of the Intel OpenVINO, a toolkit around neural networks, using its built-in benchmarking support and analyzing the throughput and latency for various models. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2022.2.devModel: Face Detection FP16 - Device: CPUmitigations=offLinux 6.1 Defaults30060090012001500SE +/- 9.54, N = 3SE +/- 8.27, N = 31557.361556.54MIN: 1398.63 / MAX: 1895.62MIN: 1401.44 / MAX: 1905.831. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -flto -shared

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2022.2.devModel: Age Gender Recognition Retail 0013 FP16 - Device: CPUmitigations=offLinux 6.1 Defaults3K6K9K12K15KSE +/- 8.91, N = 3SE +/- 5.29, N = 314683.7514690.901. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -flto -shared

Stress-NG

Stress-NG is a Linux stress tool developed by Colin King of Canonical. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: Semaphoresmitigations=offLinux 6.1 Defaults800K1600K2400K3200K4000KSE +/- 4974.11, N = 3SE +/- 6818.46, N = 33519548.923517936.221. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lrt -lsctp -lz -pthread

Cryptsetup

This is a test profile for running the cryptsetup benchmark to report on the system's cryptography performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgIterations Per Second, More Is BetterCryptsetupPBKDF2-sha512mitigations=offLinux 6.1 Defaults600K1200K1800K2400K3000KSE +/- 1281.67, N = 3SE +/- 2560.00, N = 328403872839109

OpenBenchmarking.orgMiB/s, More Is BetterCryptsetupTwofish-XTS 256b Decryptionmitigations=offLinux 6.1 Defaults150300450600750SE +/- 0.30, N = 3SE +/- 0.27, N = 3704.3704.6

Neural Magic DeepSparse

OpenBenchmarking.orgitems/sec, More Is BetterNeural Magic DeepSparse 1.1Model: CV Classification, ResNet-50 ImageNet - Scenario: Asynchronous Multi-Streammitigations=offLinux 6.1 Defaults306090120150SE +/- 1.00, N = 3SE +/- 0.64, N = 3155.32155.27

OpenVINO

This is a test of the Intel OpenVINO, a toolkit around neural networks, using its built-in benchmarking support and analyzing the throughput and latency for various models. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2022.2.devModel: Age Gender Recognition Retail 0013 FP16-INT8 - Device: CPUmitigations=offLinux 6.1 Defaults7K14K21K28K35KSE +/- 20.38, N = 3SE +/- 4.98, N = 333193.1133202.881. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -flto -shared

Cryptsetup

This is a test profile for running the cryptsetup benchmark to report on the system's cryptography performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/s, More Is BetterCryptsetupSerpent-XTS 512b Decryptionmitigations=offLinux 6.1 Defaults2004006008001000SE +/- 0.87, N = 3SE +/- 0.74, N = 31043.11043.4

OpenBenchmarking.orgMiB/s, More Is BetterCryptsetupTwofish-XTS 512b Encryptionmitigations=offLinux 6.1 Defaults150300450600750SE +/- 0.21, N = 3SE +/- 0.15, N = 3699.8699.6

Stress-NG

Stress-NG is a Linux stress tool developed by Colin King of Canonical. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: Mallocmitigations=offLinux 6.1 Defaults7M14M21M28M35MSE +/- 95450.84, N = 3SE +/- 145595.11, N = 332341790.8532335069.761. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lrt -lsctp -lz -pthread

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: x86_64 RdRandmitigations=offLinux 6.1 Defaults20K40K60K80K100KSE +/- 28.05, N = 3SE +/- 3.35, N = 382746.8782763.691. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lrt -lsctp -lz -pthread

Cryptsetup

This is a test profile for running the cryptsetup benchmark to report on the system's cryptography performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/s, More Is BetterCryptsetupTwofish-XTS 256b Encryptionmitigations=offLinux 6.1 Defaults150300450600750SE +/- 1.37, N = 3SE +/- 1.56, N = 3698.5698.4

OpenBenchmarking.orgMiB/s, More Is BetterCryptsetupTwofish-XTS 512b Decryptionmitigations=offLinux 6.1 Defaults150300450600750SE +/- 0.26, N = 3SE +/- 0.26, N = 3704.1704.2

Facebook RocksDB

OpenBenchmarking.orgOp/s, More Is BetterFacebook RocksDB 7.5.3Test: Random Readmitigations=offLinux 6.1 Defaults40M80M120M160M200MSE +/- 2519557.52, N = 3SE +/- 2292531.98, N = 31775031221774808531. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread

Neural Magic DeepSparse

OpenBenchmarking.orgms/batch, Fewer Is BetterNeural Magic DeepSparse 1.1Model: CV Classification, ResNet-50 ImageNet - Scenario: Asynchronous Multi-Streammitigations=offLinux 6.1 Defaults20406080100SE +/- 0.51, N = 3SE +/- 0.28, N = 377.2077.19

Cryptsetup

This is a test profile for running the cryptsetup benchmark to report on the system's cryptography performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMiB/s, More Is BetterCryptsetupSerpent-XTS 256b Encryptionmitigations=offLinux 6.1 Defaults2004006008001000SE +/- 2.16, N = 3SE +/- 2.59, N = 31008.91009.0

Renaissance

Renaissance is a suite of benchmarks designed to test the Java JVM from Apache Spark to a Twitter-like service to Scala and other features. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterRenaissance 0.14Test: ALS Movie Lensmitigations=offLinux 6.1 Defaults9001800270036004500SE +/- 22.27, N = 3SE +/- 32.41, N = 34371.54371.9MIN: 4345.9 / MAX: 4750.01MIN: 4308.11 / MAX: 4860.44

Dragonflydb

Dragonfly is an open-source database server that is a "modern Redis replacement" that aims to be the fastest memory store while being compliant with the Redis and Memcached protocols. For benchmarking Dragonfly, Memtier_benchmark is used as a NoSQL Redis/Memcache traffic generation plus benchmarking tool developed by Redis Labs. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgOps/sec, More Is BetterDragonflydb 0.6Clients: 200 - Set To Get Ratio: 1:5mitigations=offLinux 6.1 Defaults1.5M3M4.5M6M7.5MSE +/- 33271.78, N = 3SE +/- 41132.46, N = 37029304.507029354.711. (CXX) g++ options: -O2 -levent_openssl -levent -lcrypto -lssl -lpthread -lz -lpcre

OpenVINO

This is a test of the Intel OpenVINO, a toolkit around neural networks, using its built-in benchmarking support and analyzing the throughput and latency for various models. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2022.2.devModel: Age Gender Recognition Retail 0013 FP16-INT8 - Device: CPUmitigations=offLinux 6.1 Defaults0.1620.3240.4860.6480.81SE +/- 0.00, N = 3SE +/- 0.00, N = 30.720.72MIN: 0.5 / MAX: 5.06MIN: 0.5 / MAX: 5.071. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -flto -shared

OpenBenchmarking.orgms, Fewer Is BetterOpenVINO 2022.2.devModel: Age Gender Recognition Retail 0013 FP16 - Device: CPUmitigations=offLinux 6.1 Defaults0.36680.73361.10041.46721.834SE +/- 0.00, N = 3SE +/- 0.00, N = 31.631.63MIN: 0.83 / MAX: 10.17MIN: 1.11 / MAX: 10.181. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -flto -shared

OpenBenchmarking.orgFPS, More Is BetterOpenVINO 2022.2.devModel: Face Detection FP16 - Device: CPUmitigations=offLinux 6.1 Defaults1.15432.30863.46294.61725.7715SE +/- 0.03, N = 3SE +/- 0.03, N = 35.135.131. (CXX) g++ options: -fPIC -fsigned-char -ffunction-sections -fdata-sections -O3 -fno-strict-overflow -fwrapv -flto -shared

PostgreSQL

This is a benchmark of PostgreSQL using the integrated pgbench for facilitating the database benchmarks. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL 15Scaling Factor: 100 - Clients: 100 - Mode: Read Only - Average Latencymitigations=offLinux 6.1 Defaults0.01820.03640.05460.07280.091SE +/- 0.001, N = 4SE +/- 0.000, N = 30.0810.0811. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm

OpenBenchmarking.orgms, Fewer Is BetterPostgreSQL 15Scaling Factor: 100 - Clients: 50 - Mode: Read Only - Average Latencymitigations=offLinux 6.1 Defaults0.0090.0180.0270.0360.045SE +/- 0.000, N = 3SE +/- 0.000, N = 30.0400.0401. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lm

Apache Spark

This is a benchmark of Apache Spark with its PySpark interface. Apache Spark is an open-source unified analytics engine for large-scale data processing and dealing with big data. This test profile benchmars the Apache Spark in a single-system configuration using spark-submit. The test makes use of DIYBigData's pyspark-benchmark (https://github.com/DIYBigData/pyspark-benchmark/) for generating of test data and various Apache Spark operations. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterApache Spark 3.3Row Count: 1000000 - Partitions: 100 - Group By Test Timemitigations=offLinux 6.1 Defaults0.57381.14761.72142.29522.869SE +/- 0.01, N = 15SE +/- 0.02, N = 152.552.55

OpenBenchmarking.orgSeconds, Fewer Is BetterApache Spark 3.3Row Count: 1000000 - Partitions: 100 - SHA-512 Benchmark Timemitigations=offLinux 6.1 Defaults0.44780.89561.34341.79122.239SE +/- 0.02, N = 15SE +/- 0.02, N = 151.991.99

srsRAN

srsRAN is an open-source LTE/5G software radio suite created by Software Radio Systems (SRS). The srsRAN radio suite was formerly known as srsLTE and can be used for building your own software-defined radio (SDR) 4G/5G mobile network. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgeNb Mb/s, More Is BettersrsRAN 22.04.1Test: 4G PHY_DL_Test 100 PRB SISO 256-QAMmitigations=offLinux 6.1 Defaults150300450600750SE +/- 1.00, N = 3SE +/- 7.34, N = 3713.6713.61. (CXX) g++ options: -std=c++14 -fno-strict-aliasing -march=native -mfpmath=sse -mavx2 -fvisibility=hidden -O3 -fno-trapping-math -fno-math-errno -ldl -lpthread -lm

WebP2 Image Encode

This is a test of Google's libwebp2 library with the WebP2 image encode utility and using a sample 6000x4000 pixel JPEG image as the input, similar to the WebP/libwebp test profile. WebP2 is currently experimental and under heavy development as ultimately the successor to WebP. WebP2 supports 10-bit HDR, more efficienct lossy compression, improved lossless compression, animation support, and full multi-threading support compared to WebP. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMP/s, More Is BetterWebP2 Image Encode 20220823Encode Settings: Quality 100, Lossless Compressionmitigations=offLinux 6.1 Defaults0.0090.0180.0270.0360.045SE +/- 0.00, N = 3SE +/- 0.00, N = 30.040.041. (CXX) g++ options: -msse4.2 -fno-rtti -O3 -ldl

OpenBenchmarking.orgMP/s, More Is BetterWebP2 Image Encode 20220823Encode Settings: Quality 95, Compression Effort 7mitigations=offLinux 6.1 Defaults0.0360.0720.1080.1440.18SE +/- 0.00, N = 3SE +/- 0.00, N = 30.160.161. (CXX) g++ options: -msse4.2 -fno-rtti -O3 -ldl

OpenBenchmarking.orgMP/s, More Is BetterWebP2 Image Encode 20220823Encode Settings: Quality 75, Compression Effort 7mitigations=offLinux 6.1 Defaults0.07430.14860.22290.29720.3715SE +/- 0.00, N = 3SE +/- 0.00, N = 30.330.331. (CXX) g++ options: -msse4.2 -fno-rtti -O3 -ldl

WebP Image Encode

This is a test of Google's libwebp with the cwebp image encode utility and using a sample 6000x4000 pixel JPEG image as the input. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMP/s, More Is BetterWebP Image Encode 1.2.4Encode Settings: Quality 100, Highest Compressionmitigations=offLinux 6.1 Defaults1.16552.3313.49654.6625.8275SE +/- 0.00, N = 3SE +/- 0.00, N = 35.185.181. (CC) gcc options: -fvisibility=hidden -O2 -lm

OpenBenchmarking.orgMP/s, More Is BetterWebP Image Encode 1.2.4Encode Settings: Quality 100, Losslessmitigations=offLinux 6.1 Defaults0.52651.0531.57952.1062.6325SE +/- 0.00, N = 3SE +/- 0.00, N = 32.342.341. (CC) gcc options: -fvisibility=hidden -O2 -lm

Facebook RocksDB

OpenBenchmarking.orgOp/s, More Is BetterFacebook RocksDB 7.5.3Test: Read While Writingmitigations=offLinux 6.1 Defaults1.1M2.2M3.3M4.4M5.5MSE +/- 97717.39, N = 15SE +/- 97089.51, N = 15493058150898871. (CXX) g++ options: -O3 -march=native -pthread -fno-builtin-memcmp -fno-rtti -lpthread

Stress-NG

Stress-NG is a Linux stress tool developed by Colin King of Canonical. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: Context Switchingmitigations=offLinux 6.1 Defaults3M6M9M12M15MSE +/- 538579.50, N = 12SE +/- 457851.80, N = 1216210513.9711794492.431. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lrt -lsctp -lz -pthread

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: CPU Cachemitigations=offLinux 6.1 Defaults20406080100SE +/- 1.99, N = 15SE +/- 1.56, N = 1599.4297.761. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lrt -lsctp -lz -pthread

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: Atomicmitigations=offLinux 6.1 Defaults70K140K210K280K350KSE +/- 2320.39, N = 3SE +/- 7379.51, N = 15331842.59338549.561. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lrt -lsctp -lz -pthread

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.14.06Test: Mutexmitigations=offLinux 6.1 Defaults4M8M12M16M20MSE +/- 153223.16, N = 3SE +/- 368734.03, N = 1217479306.2716677300.351. (CC) gcc options: -O2 -std=gnu99 -lm -fuse-ld=gold -lapparmor -latomic -lc -lcrypt -ldl -ljpeg -lrt -lsctp -lz -pthread

Apache Spark

This is a benchmark of Apache Spark with its PySpark interface. Apache Spark is an open-source unified analytics engine for large-scale data processing and dealing with big data. This test profile benchmars the Apache Spark in a single-system configuration using spark-submit. The test makes use of DIYBigData's pyspark-benchmark (https://github.com/DIYBigData/pyspark-benchmark/) for generating of test data and various Apache Spark operations. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterApache Spark 3.3Row Count: 1000000 - Partitions: 500 - Broadcast Inner Join Test Timemitigations=offLinux 6.1 Defaults0.17780.35560.53340.71120.889SE +/- 0.02, N = 7SE +/- 0.03, N = 30.780.79

OpenBenchmarking.orgSeconds, Fewer Is BetterApache Spark 3.3Row Count: 1000000 - Partitions: 100 - Broadcast Inner Join Test Timemitigations=offLinux 6.1 Defaults0.18450.3690.55350.7380.9225SE +/- 0.01, N = 15SE +/- 0.02, N = 150.760.82

OpenBenchmarking.orgSeconds, Fewer Is BetterApache Spark 3.3Row Count: 1000000 - Partitions: 100 - Inner Join Test Timemitigations=offLinux 6.1 Defaults0.20930.41860.62790.83721.0465SE +/- 0.01, N = 15SE +/- 0.02, N = 150.870.93

WebP2 Image Encode

This is a test of Google's libwebp2 library with the WebP2 image encode utility and using a sample 6000x4000 pixel JPEG image as the input, similar to the WebP/libwebp test profile. WebP2 is currently experimental and under heavy development as ultimately the successor to WebP. WebP2 supports 10-bit HDR, more efficienct lossy compression, improved lossless compression, animation support, and full multi-threading support compared to WebP. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMP/s, More Is BetterWebP2 Image Encode 20220823Encode Settings: Quality 100, Compression Effort 5mitigations=offLinux 6.1 Defaults3691215SE +/- 0.06, N = 3SE +/- 0.16, N = 159.469.111. (CXX) g++ options: -msse4.2 -fno-rtti -O3 -ldl

DaCapo Benchmark

This test runs the DaCapo Benchmarks written in Java and intended to test system/CPU performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgmsec, Fewer Is BetterDaCapo Benchmark 9.12-MR1Java Test: H2mitigations=offLinux 6.1 Defaults5001000150020002500SE +/- 52.45, N = 20SE +/- 41.36, N = 2021372111

198 Results Shown

Renaissance
Apache HBase:
  1000000 - Rand Read - 1:
    Rows Per Second
    Microseconds - Average Latency
  1000000 - Rand Read - 16:
    Microseconds - Average Latency
Apache CouchDB
Stress-NG
ctx_clock
OSBench
Apache HBase
libavif avifenc
Stress-NG
AOM AV1
libavif avifenc
Apache HBase
AOM AV1
Renaissance:
  In-Memory Database Shootout
  Akka Unbalanced Cobwebbed Tree
Stress-NG
srsRAN
AOM AV1
Apache CouchDB
srsRAN
Renaissance
Apache Spark
libavif avifenc
Renaissance
WebP2 Image Encode
Apache HBase
OSBench
Renaissance
Dragonflydb
ClickHouse
Stress-NG
RawTherapee
Dragonflydb
Stress-NG
SVT-AV1
Neural Magic DeepSparse:
  NLP Text Classification, BERT base uncased SST2 - Asynchronous Multi-Stream:
    ms/batch
    items/sec
Renaissance
Stress-NG
SVT-AV1
Apache Spark
AOM AV1
srsRAN
Stress-NG
Neural Magic DeepSparse
ASTC Encoder
AOM AV1
WebP Image Encode
Apache HBase
Neural Magic DeepSparse
ClickHouse
Apache HBase
Stress-NG
OpenVINO:
  Person Vehicle Bike Detection FP16 - CPU:
    FPS
    ms
Apache Spark
OSBench
Stress-NG
DaCapo Benchmark
Facebook RocksDB:
  Rand Fill Sync
  Rand Fill
srsRAN
MariaDB
srsRAN
KeyDB
Facebook RocksDB
Apache CouchDB
Renaissance
Dragonflydb
OpenVINO:
  Weld Porosity Detection FP16-INT8 - CPU:
    ms
    FPS
Stress-NG
ASTC Encoder
srsRAN
PostgreSQL:
  100 - 50 - Read Only
  100 - 100 - Read Only
InfluxDB
srsRAN
Stress-NG
Neural Magic DeepSparse
Apache Spark
DaCapo Benchmark
Neural Magic DeepSparse
Bork File Encrypter
Renaissance
OpenVINO:
  Person Detection FP32 - CPU:
    ms
    FPS
Neural Magic DeepSparse
OpenVINO
Neural Magic DeepSparse
Facebook RocksDB
OpenVINO
Neural Magic DeepSparse
SVT-AV1
OpenVINO
Neural Magic DeepSparse:
  NLP Token Classification, BERT base uncased conll2003 - Asynchronous Multi-Stream
  NLP Question Answering, BERT base uncased SQuaD 12layer Pruned90 - Asynchronous Multi-Stream
OpenVINO
Facebook RocksDB
OSBench
libavif avifenc:
  0
  2
SVT-AV1
Stargate Digital Audio Workstation
Cryptsetup
srsRAN
Aircrack-ng
Stress-NG:
  Memory Copying
  Crypto
Cryptsetup
DaCapo Benchmark
Dragonflydb
AOM AV1
Renaissance
ClickHouse
WebP Image Encode
Neural Magic DeepSparse
OpenVINO
Apache Spark
SQLite Speedtest
nginx
Stargate Digital Audio Workstation
Cryptsetup
WebP Image Encode
Java Gradle Build
OpenVINO:
  Weld Porosity Detection FP16 - CPU:
    FPS
    ms
nginx
Cryptsetup
nginx
Cryptsetup
srsRAN
OpenVINO
InfluxDB
Timed Gem5 Compilation
Cryptsetup
OpenVINO:
  Machine Translation EN To DE FP16 - CPU
  Person Detection FP16 - CPU
Stress-NG
nginx
OpenVINO
OSBench
Timed Linux Kernel Compilation
Stress-NG
ASTC Encoder
srsRAN
AOM AV1
BRL-CAD
ASTC Encoder
Apache CouchDB
Timed Linux Kernel Compilation
Cryptsetup
OpenVINO
Apache Spark
Dragonflydb
OpenVINO:
  Face Detection FP16 - CPU
  Age Gender Recognition Retail 0013 FP16 - CPU
Stress-NG
Cryptsetup:
  PBKDF2-sha512
  Twofish-XTS 256b Decryption
Neural Magic DeepSparse
OpenVINO
Cryptsetup:
  Serpent-XTS 512b Decryption
  Twofish-XTS 512b Encryption
Stress-NG:
  Malloc
  x86_64 RdRand
Cryptsetup:
  Twofish-XTS 256b Encryption
  Twofish-XTS 512b Decryption
Facebook RocksDB
Neural Magic DeepSparse
Cryptsetup
Renaissance
Dragonflydb
OpenVINO:
  Age Gender Recognition Retail 0013 FP16-INT8 - CPU
  Age Gender Recognition Retail 0013 FP16 - CPU
  Face Detection FP16 - CPU
PostgreSQL:
  100 - 100 - Read Only - Average Latency
  100 - 50 - Read Only - Average Latency
Apache Spark:
  1000000 - 100 - Group By Test Time
  1000000 - 100 - SHA-512 Benchmark Time
srsRAN
WebP2 Image Encode:
  Quality 100, Lossless Compression
  Quality 95, Compression Effort 7
  Quality 75, Compression Effort 7
WebP Image Encode:
  Quality 100, Highest Compression
  Quality 100, Lossless
Facebook RocksDB
Stress-NG:
  Context Switching
  CPU Cache
  Atomic
  Mutex
Apache Spark:
  1000000 - 500 - Broadcast Inner Join Test Time
  1000000 - 100 - Broadcast Inner Join Test Time
  1000000 - 100 - Inner Join Test Time
WebP2 Image Encode
DaCapo Benchmark

Linux 6.1 Defaults

Kernel Notes: Transparent Huge Pages: madvise
Compiler Notes: --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,brig,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-link-serialization=2 --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-11-xKiWfi/gcc-11-11.3.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-11-xKiWfi/gcc-11-11.3.0/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
Processor Notes: Scaling Governor: intel_pstate powersave (EPP: balance_performance) - CPU Microcode: 0x10e - Thermald 2.4.9
Java Notes: OpenJDK Runtime Environment (build 11.0.17+8-post-Ubuntu-1ubuntu222.04)
Python Notes: Python 3.10.6
Security Notes: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: 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 IBRS IBPB: conditional RSB filling PBRSB-eIBRS: SW sequence + srbds: Not affected + tsx_async_abort: Not affected

Testing initiated at 28 November 2022 12:55 by user phoronix.

mitigations=off

Processor: Intel Core i9-13900K @ 5.50GHz (24 Cores / 32 Threads), Motherboard: ASUS PRIME Z790-P WIFI (0602 BIOS), Chipset: Intel Device 7a27, Memory: 32GB, Disk: 1000GB Western Digital WDS100T1X0E-00AFY0, Graphics: AMD Radeon RX 6800 16GB (2475/1000MHz), Audio: Realtek ALC897, Monitor: ASUS VP28U, Network: Realtek RTL8125 2.5GbE + Intel Device 7a70

OS: Ubuntu 22.04, Kernel: 6.1.0-060100rc5daily20221117-generic (x86_64), Desktop: GNOME Shell 42.5, Display Server: X Server 1.21.1.3 + Wayland, OpenGL: 4.6 Mesa 22.0.5 (LLVM 13.0.1 DRM 3.49), Vulkan: 1.3.204, Compiler: GCC 11.3.0, File-System: ext4, Screen Resolution: 3840x2160

Kernel Notes: Transparent Huge Pages: madvise
Compiler Notes: --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,brig,d,fortran,objc,obj-c++,m2 --enable-libphobos-checking=release --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-link-serialization=2 --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none=/build/gcc-11-xKiWfi/gcc-11-11.3.0/debian/tmp-nvptx/usr,amdgcn-amdhsa=/build/gcc-11-xKiWfi/gcc-11-11.3.0/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
Processor Notes: Scaling Governor: intel_pstate powersave (EPP: balance_performance) - CPU Microcode: 0x10e - Thermald 2.4.9
Java Notes: OpenJDK Runtime Environment (build 11.0.17+8-post-Ubuntu-1ubuntu222.04)
Python Notes: Python 3.10.6
Security Notes: itlb_multihit: Not affected + l1tf: Not affected + mds: Not affected + meltdown: Not affected + mmio_stale_data: Not affected + retbleed: Not affected + spec_store_bypass: Vulnerable + spectre_v1: Vulnerable: __user pointer sanitization and usercopy barriers only; no swapgs barriers + spectre_v2: Vulnerable IBPB: disabled STIBP: disabled PBRSB-eIBRS: Vulnerable + srbds: Not affected + tsx_async_abort: Not affected

Testing initiated at 29 November 2022 05:15 by user phoronix.