Xeon Cascade Lake R Intel FSGSBASE

Intel FSGSBASE benchmarking 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 2006246-NE-XEONGOLD517
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

Limit displaying results to tests within:

AV1 3 Tests
Bioinformatics 4 Tests
C++ Boost Tests 2 Tests
Timed Code Compilation 4 Tests
C/C++ Compiler Tests 15 Tests
CPU Massive 25 Tests
Creator Workloads 6 Tests
Database Test Suite 9 Tests
Disk Test Suite 2 Tests
Encoding 4 Tests
HPC - High Performance Computing 12 Tests
Java 2 Tests
Common Kernel Benchmarks 7 Tests
Machine Learning 4 Tests
Molecular Dynamics 4 Tests
MPI Benchmarks 3 Tests
Multi-Core 17 Tests
NVIDIA GPU Compute 2 Tests
OpenMPI Tests 3 Tests
Programmer / Developer System Benchmarks 4 Tests
Python 2 Tests
Scientific Computing 8 Tests
Server 13 Tests
Server CPU Tests 15 Tests
Single-Threaded 2 Tests
Video Encoding 4 Tests

Statistics

Show Overall Harmonic Mean(s)
Show Overall Geometric Mean
Show Geometric Means Per-Suite/Category
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
Hide
Result
Result
Identifier
View Logs
Performance Per
Dollar
Date
Run
  Test
  Duration
nofsgsbase
June 22 2020
  23 Hours, 32 Minutes
FSGSBASE Enabled
June 20 2020
  1 Day, 2 Hours, 10 Minutes
Invert Hiding All Results Option
  1 Day, 51 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):

FSGSBASE Enabled

Environment Notes: CXXFLAGS="-O3 -march=native" CFLAGS="-O3 -march=native"
Compiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --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++,gm2 --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none,hsa --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-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v
Disk Notes: MQ-DEADLINE / errors=remount-ro,relatime,rw
Processor Notes: Scaling Governor: intel_pstate powersave - CPU Microcode: 0x500002c
Java Notes: OpenJDK Runtime Environment (build 11.0.7-ea+9-post-Ubuntu-1ubuntu1)
Python Notes: Python 3.8.2
Security Notes: itlb_multihit: KVM: Mitigation of Split huge pages + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced IBRS IBPB: conditional RSB filling + srbds: Not affected + tsx_async_abort: Mitigation of TSX disabled

nofsgsbase

Processor: 2 x Intel Xeon Gold 5220R @ 3.90GHz (36 Cores / 72 Threads), Motherboard: TYAN S7106 (V2.01.B40 BIOS), Chipset: Intel Sky Lake-E DMI3 Registers, Memory: 94GB, Disk: 500GB Samsung SSD 860, Graphics: ASPEED, Monitor: VE228, Network: 2 x Intel I210 + 2 x QLogic cLOM8214 1/10GbE

OS: Ubuntu 20.04, Kernel: 5.8.0-rc1-phx-fsgsbase (x86_64) 20200620, Desktop: GNOME Shell 3.36.1, Display Server: X Server 1.20.8, Display Driver: modesetting 1.20.8, Compiler: GCC 9.3.0, File-System: ext4, Screen Resolution: 1920x1080

Environment Notes: CXXFLAGS="-O3 -march=native" CFLAGS="-O3 -march=native"
Compiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --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++,gm2 --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none,hsa --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-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v
Disk Notes: MQ-DEADLINE / errors=remount-ro,relatime,rw
Processor Notes: Scaling Governor: intel_pstate powersave - CPU Microcode: 0x5002f01
Java Notes: OpenJDK Runtime Environment (build 11.0.7+10-post-Ubuntu-3ubuntu1)
Python Notes: Python 3.8.2
Security Notes: itlb_multihit: KVM: Mitigation of Split huge pages + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced IBRS IBPB: conditional RSB filling + srbds: Not affected + tsx_async_abort: Mitigation of TSX disabled

Xeon Cascade Lake R Intel FSGSBASEOpenBenchmarking.orgPhoronix Test Suite2 x Intel Xeon Gold 5220R @ 3.90GHz (36 Cores / 72 Threads)TYAN S7106 (V2.01.B40 BIOS)Intel Sky Lake-E DMI3 Registers94GB500GB Samsung SSD 860ASPEEDVE2282 x Intel I210 + 2 x QLogic cLOM8214 1/10GbEUbuntu 20.045.8.0-rc1-phx-fsgsbase (x86_64) 20200620GNOME Shell 3.36.1X Server 1.20.8modesetting 1.20.8GCC 9.3.0ext41920x1080ProcessorMotherboardChipsetMemoryDiskGraphicsMonitorNetworkOSKernelDesktopDisplay ServerDisplay DriverCompilerFile-SystemScreen ResolutionXeon Cascade Lake R Intel FSGSBASE BenchmarksSystem Logs- CXXFLAGS="-O3 -march=native" CFLAGS="-O3 -march=native"- --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --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++,gm2 --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none,hsa --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-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 - MQ-DEADLINE / errors=remount-ro,relatime,rw- FSGSBASE Enabled: Scaling Governor: intel_pstate powersave - CPU Microcode: 0x500002c - nofsgsbase: Scaling Governor: intel_pstate powersave - CPU Microcode: 0x5002f01- FSGSBASE Enabled: OpenJDK Runtime Environment (build 11.0.7-ea+9-post-Ubuntu-1ubuntu1) - nofsgsbase: OpenJDK Runtime Environment (build 11.0.7+10-post-Ubuntu-3ubuntu1)- Python 3.8.2- itlb_multihit: KVM: Mitigation of Split huge pages + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced IBRS IBPB: conditional RSB filling + srbds: Not affected + tsx_async_abort: Mitigation of TSX disabled

FSGSBASE Enabled vs. nofsgsbase ComparisonPhoronix Test SuiteBaseline+19.9%+19.9%+39.8%+39.8%+59.7%+59.7%7.5%3.6%3.1%2.1%2%2%Buffer Test - Heavy Contention - Read Write79.5%Buffer Test - Normal Load - Read Write77.7%LPOP52.9%Rand Write - IO_uring - Yes - No - 4KB40.1%Rand Write - IO_uring - Yes - No - 2MB38.5%Context Switching19.9%Write18.5%51215%Rand Fill Sync15%Savina Reactors.IO11.8%20011.5%Seq Write - IO_uring - Yes - No - 2MB11.3%Seq Write - IO_uring - Yes - No - 2MB11.2%Writes9.5%Redis8.5%Fayalite-FIST DataGET6.9%Increment - 15.8%Increment - 15.6%Atomic4.9%2564.9%Rhodopsin Protein4.7%T.T.F.S.S4.4%Reactor4.4%No - Inference - VGG19 - CPUSeq Read - 13.5%1283.2%Seq Read - 13.2%Seq FillApache Spark ALS3.1%643%scikit_linearridgeregression2.6%Speed 6 Two-Pass2.4%2.4%Async Rand Read - 12.1%Async Rand Read - 12.1%Seq FillNo - Inference - IMDB LSTM - CPU2.1%Rand Fill72 - 100% ReadsPostgreSQL pgbenchPostgreSQL pgbenchRedisFlexible IO TesterFlexible IO TesterStress-NGBlogBenchMariaDBFacebook RocksDBRenaissanceApache SiegeFlexible IO TesterFlexible IO TesterApache CassandraMemtier_benchmarkCP2K Molecular DynamicsRedisApache HBaseApache HBaseStress-NGMariaDBLAMMPS Molecular Dynamics SimulatorYafaRayJava Gradle BuildPlaidMLApache HBaseMariaDBApache HBaseLevelDBRenaissanceMariaDBMlpack BenchmarkAOM AV1KeyDBApache HBaseApache HBaseLevelDBPlaidMLLevelDBpmbenchFSGSBASE Enablednofsgsbase

Xeon Cascade Lake R Intel FSGSBASEaom-av1: Speed 0 Two-Passaom-av1: Speed 4 Two-Passaom-av1: Speed 6 Realtimeaom-av1: Speed 6 Two-Passaom-av1: Speed 8 Realtimecassandra: Writeshbase: Increment - 1hbase: Increment - 1hbase: Rand Read - 1hbase: Rand Read - 1hbase: Seq Read - 1hbase: Seq Read - 1hbase: Async Rand Read - 1hbase: Async Rand Read - 1apache-siege: 10apache-siege: 50apache-siege: 200blogbench: Writecp2k: Fayalite-FIST Datactx-clock: Context Switch Timedav1d: Chimera 1080pdav1d: Summer Nature 4Kdav1d: Summer Nature 1080pdav1d: Chimera 1080p 10-bitebizzy: rocksdb: Rand Fillrocksdb: Rand Readrocksdb: Seq Fillrocksdb: Rand Fill Syncrocksdb: Read While Writingfio: Rand Write - IO_uring - Yes - No - 2MB - Default Test Directoryfio: Rand Write - IO_uring - Yes - No - 4KB - Default Test Directoryfio: Seq Write - IO_uring - Yes - No - 2MB - Default Test Directoryfio: Seq Write - IO_uring - Yes - No - 2MB - Default Test Directorygromacs: Water Benchmarkhimeno: Poisson Pressure Solverjava-gradle-perf: Reactorkeydb: lammps: Rhodopsin Proteinleveldb: Hot Readleveldb: Fill Syncleveldb: Fill Syncleveldb: Overwriteleveldb: Overwriteleveldb: Rand Fillleveldb: Rand Fillleveldb: Rand Readleveldb: Seek Randleveldb: Rand Deleteleveldb: Seq Fillleveldb: Seq Fillmysqlslap: 64mysqlslap: 128mysqlslap: 256mysqlslap: 512memtier-benchmark: Redismlpack: scikit_icamlpack: scikit_qdamlpack: scikit_svmmlpack: scikit_linearridgeregressionnamd: ATPase Simulation - 327,506 Atomsnode-express-loadtest: numenta-nab: EXPoSEnumenta-nab: Relative Entropynumenta-nab: Earthgecko Skylinenumenta-nab: Bayesian Changepointonednn: IP Batch 1D - bf16bf16bf16 - CPUonednn: IP Batch All - bf16bf16bf16 - CPUonednn: Convolution Batch Shapes Auto - bf16bf16bf16 - CPUonednn: Deconvolution Batch deconv_1d - bf16bf16bf16 - CPUonednn: Deconvolution Batch deconv_3d - bf16bf16bf16 - CPUonednn: Matrix Multiply Batch Shapes Transformer - bf16bf16bf16 - CPUplaidml: No - Inference - VGG16 - CPUplaidml: No - Inference - VGG19 - CPUplaidml: No - Inference - IMDB LSTM - CPUplaidml: No - Inference - Mobilenet - CPUplaidml: No - Inference - ResNet 50 - CPUplaidml: No - Inference - DenseNet 201 - CPUplaidml: No - Inference - Inception V3 - CPUplaidml: No - Inference - NASNer Large - CPUpmbench: 72 - 100% Readspmbench: 72 - 100% Writespmbench: 1 - 80% Reads 20% Writespgbench: Buffer Test - Normal Load - Read Onlypgbench: Buffer Test - Normal Load - Read Writepgbench: Buffer Test - Heavy Contention - Read Onlypgbench: Buffer Test - Heavy Contention - Read Writepostmark: Disk Transaction Performanceqmcpack: redis: LPOPredis: SADDredis: GETredis: SETrenaissance: Apache Spark ALSrenaissance: Savina Reactors.IOstress-ng: Atomicstress-ng: SENDFILEstress-ng: CPU Stressstress-ng: Context Switchingsvt-av1: Enc Mode 0 - 1080psvt-av1: Enc Mode 4 - 1080psvt-av1: Enc Mode 8 - 1080pbuild-apache: Time To Compilebuild-gdb: Time To Compilehmmer: Pfam Database Searchbuild-linux-kernel: Time To Compilebuild-llvm: Time To Compilemafft: Multiple Sequence Alignmentvpxenc: Speed 0vpxenc: Speed 5yafaray: Total Time For Sample SceneFSGSBASE Enablednofsgsbase0.271.9510.862.9823.74147023341329146432135270189524518922712.0533173.4548540.41242472027.693124329.39182.78338.3687.20108718318609314220564818795065325356272187888003851893.5063932.762615276.580428585.0117.95092.2771.84485.3509.9799.3569.8809.69593.999113.581774.3579.6826.2682051591501612859981.4074.6646.0328.711.960.6110493951513.77914.81390.77532.7385.6956251.09646.397287.391549.461581.4492324.9820.64868.0910.624.451.965.500.570.04600.08020.0756593329.4276834908.501704618994.7427154727.71374457252687.62611119.752087688.752500801.501918164.832896.19121837.00390016.18447930.4711896.429410762.670.1205.77048.87425.764137.39812.71937.431283.7372.6606.1423.01108.8180.271.9410.782.9123.84134285323130846252145090195513719322746.8833180.6643531.70204551886.257125328.03180.77335.1687.47108265018645114144819818964356815396411135634003461703.5143966.355979288.793418712.1017.14191.6461.84460.30810.0792.43910.0798.60193.010113.203761.4349.9809.4411991541431402635763.8275.1246.0928.462.010.6107793421500.84714.91389.60932.6015.6791051.11346.387357.390639.461631.4519325.1521.39850.6510.604.481.985.450.580.04510.08120.0756594395.9066912762.902255619694.8556372634.26631957252688.51707562.792071005.922340376.21908415.462985.59624407.08385800.80444432.1011983.667847877.590.125.68349.21125.474135.38412.72737.181285.3002.6656.1223.24113.629OpenBenchmarking.org

AOM AV1

This is a simple test of the AOMedia AV1 encoder run on the CPU with a sample video file. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 2.0Encoder Mode: Speed 0 Two-PassFSGSBASE Enablednofsgsbase0.06080.12160.18240.24320.304SE +/- 0.00, N = 3SE +/- 0.00, N = 30.270.271. (CXX) g++ options: -O3 -march=native -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 2.0Encoder Mode: Speed 4 Two-PassFSGSBASE Enablednofsgsbase0.43880.87761.31641.75522.194SE +/- 0.00, N = 3SE +/- 0.00, N = 31.951.941. (CXX) g++ options: -O3 -march=native -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 2.0Encoder Mode: Speed 6 RealtimeFSGSBASE Enablednofsgsbase3691215SE +/- 0.08, N = 3SE +/- 0.09, N = 310.8610.781. (CXX) g++ options: -O3 -march=native -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 2.0Encoder Mode: Speed 6 Two-PassFSGSBASE Enablednofsgsbase0.67051.3412.01152.6823.3525SE +/- 0.00, N = 3SE +/- 0.02, N = 32.982.911. (CXX) g++ options: -O3 -march=native -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread

OpenBenchmarking.orgFrames Per Second, More Is BetterAOM AV1 2.0Encoder Mode: Speed 8 RealtimeFSGSBASE Enablednofsgsbase612182430SE +/- 0.17, N = 3SE +/- 0.31, N = 323.7423.841. (CXX) g++ options: -O3 -march=native -std=c++11 -U_FORTIFY_SOURCE -lm -lpthread

Apache Cassandra

This is a benchmark of the Apache Cassandra NoSQL database management system making use of cassandra-stress. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgOp/s, More Is BetterApache Cassandra 3.11.4Test: WritesFSGSBASE Enablednofsgsbase30K60K90K120K150KSE +/- 2666.43, N = 15SE +/- 1934.98, N = 15147023134285

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 2.2.3Test: Increment - Clients: 1FSGSBASE Enablednofsgsbase7001400210028003500SE +/- 30.10, N = 11SE +/- 28.49, N = 1534133231

OpenBenchmarking.orgMicroseconds - Average Latency, Fewer Is BetterApache HBase 2.2.3Test: Increment - Clients: 1FSGSBASE Enablednofsgsbase70140210280350SE +/- 2.52, N = 11SE +/- 2.70, N = 15291308

OpenBenchmarking.orgRows Per Second, More Is BetterApache HBase 2.2.3Test: Random Read - Clients: 1FSGSBASE Enablednofsgsbase10002000300040005000SE +/- 56.05, N = 15SE +/- 48.14, N = 1546434625

OpenBenchmarking.orgMicroseconds - Average Latency, Fewer Is BetterApache HBase 2.2.3Test: Random Read - Clients: 1FSGSBASE Enablednofsgsbase50100150200250SE +/- 2.57, N = 15SE +/- 2.20, N = 15213214

OpenBenchmarking.orgRows Per Second, More Is BetterApache HBase 2.2.3Test: Sequential Read - Clients: 1FSGSBASE Enablednofsgsbase11002200330044005500SE +/- 93.78, N = 15SE +/- 44.45, N = 1552705090

OpenBenchmarking.orgMicroseconds - Average Latency, Fewer Is BetterApache HBase 2.2.3Test: Sequential Read - Clients: 1FSGSBASE Enablednofsgsbase4080120160200SE +/- 3.62, N = 15SE +/- 1.75, N = 15189195

OpenBenchmarking.orgRows Per Second, More Is BetterApache HBase 2.2.3Test: Async Random Read - Clients: 1FSGSBASE Enablednofsgsbase11002200330044005500SE +/- 81.85, N = 15SE +/- 78.77, N = 1252455137

OpenBenchmarking.orgMicroseconds - Average Latency, Fewer Is BetterApache HBase 2.2.3Test: Async Random Read - Clients: 1FSGSBASE Enablednofsgsbase4080120160200SE +/- 3.20, N = 15SE +/- 3.36, N = 12189193

Apache Siege

This is a test of the Apache web server performance being facilitated by the Siege web serverb enchmark program. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgTransactions Per Second, More Is BetterApache Siege 2.4.29Concurrent Users: 10FSGSBASE Enablednofsgsbase5K10K15K20K25KSE +/- 150.23, N = 3SE +/- 164.71, N = 322712.0522746.881. (CC) gcc options: -O3 -march=native -lpthread -ldl -lssl -lcrypto

OpenBenchmarking.orgTransactions Per Second, More Is BetterApache Siege 2.4.29Concurrent Users: 50FSGSBASE Enablednofsgsbase7K14K21K28K35KSE +/- 194.24, N = 3SE +/- 188.88, N = 333173.4533180.661. (CC) gcc options: -O3 -march=native -lpthread -ldl -lssl -lcrypto

OpenBenchmarking.orgTransactions Per Second, More Is BetterApache Siege 2.4.29Concurrent Users: 200FSGSBASE Enablednofsgsbase10K20K30K40K50KSE +/- 1308.05, N = 12SE +/- 254.67, N = 348540.4143531.701. (CC) gcc options: -O3 -march=native -lpthread -ldl -lssl -lcrypto

BlogBench

BlogBench is designed to replicate the load of a real-world busy file server by stressing the file-system with multiple threads of random reads, writes, and rewrites. The behavior is mimicked of that of a blog by creating blogs with content and pictures, modifying blog posts, adding comments to these blogs, and then reading the content of the blogs. All of these blogs generated are created locally with fake content and pictures. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFinal Score, More Is BetterBlogBench 1.1Test: WriteFSGSBASE Enablednofsgsbase5K10K15K20K25KSE +/- 920.31, N = 3SE +/- 1371.68, N = 324247204551. (CC) gcc options: -O3 -march=native -pthread

CP2K Molecular Dynamics

CP2K is an open-source molecular dynamics software package focused on quantum chemistry and solid-state physics. This test profile currently makes use of the OpenMP implementation and using the Fayalite-FIST molecular dynamics run and measures the total time to complete. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterCP2K Molecular Dynamics 6.1Fayalite-FIST DataFSGSBASE Enablednofsgsbase4008001200160020002027.691886.26

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 TimeFSGSBASE Enablednofsgsbase306090120150SE +/- 0.67, N = 31241251. (CC) gcc options: -O3 -march=native

dav1d

Dav1d is an open-source, speedy AV1 video decoder. This test profile times how long it takes to decode sample AV1 video content. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is Betterdav1d 0.7.0Video Input: Chimera 1080pFSGSBASE Enablednofsgsbase70140210280350SE +/- 3.13, N = 3SE +/- 4.08, N = 3329.39328.03MIN: 204.26 / MAX: 425.36MIN: 183.84 / MAX: 426.681. (CC) gcc options: -O3 -march=native -pthread

OpenBenchmarking.orgFPS, More Is Betterdav1d 0.7.0Video Input: Summer Nature 4KFSGSBASE Enablednofsgsbase4080120160200SE +/- 2.74, N = 3SE +/- 0.93, N = 3182.78180.77MIN: 88.23 / MAX: 199.52MIN: 91.75 / MAX: 195.311. (CC) gcc options: -O3 -march=native -pthread

OpenBenchmarking.orgFPS, More Is Betterdav1d 0.7.0Video Input: Summer Nature 1080pFSGSBASE Enablednofsgsbase70140210280350SE +/- 1.19, N = 3SE +/- 1.40, N = 3338.36335.16MIN: 185.24 / MAX: 374.84MIN: 172.66 / MAX: 372.41. (CC) gcc options: -O3 -march=native -pthread

OpenBenchmarking.orgFPS, More Is Betterdav1d 0.7.0Video Input: Chimera 1080p 10-bitFSGSBASE Enablednofsgsbase20406080100SE +/- 0.13, N = 3SE +/- 0.10, N = 387.2087.47MIN: 66.61 / MAX: 133.73MIN: 66.73 / MAX: 137.931. (CC) gcc options: -O3 -march=native -pthread

ebizzy

This is a test of ebizzy, a program to generate workloads resembling web server workloads. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgRecords/s, More Is Betterebizzy 0.3FSGSBASE Enablednofsgsbase200K400K600K800K1000KSE +/- 10256.56, N = 15SE +/- 9309.86, N = 3108718310826501. (CC) gcc options: -pthread -lpthread -O3 -march=native

Facebook RocksDB

This is a benchmark of Facebook's RocksDB as an embeddable persistent key-value store for fast storage based on Google's LevelDB. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgOp/s, More Is BetterFacebook RocksDB 6.3.6Test: Random FillFSGSBASE Enablednofsgsbase40K80K120K160K200KSE +/- 172.09, N = 3SE +/- 226.90, N = 31860931864511. (CXX) g++ options: -O3 -march=native -std=c++11 -fno-builtin-memcmp -fno-rtti -rdynamic -lpthread

OpenBenchmarking.orgOp/s, More Is BetterFacebook RocksDB 6.3.6Test: Random ReadFSGSBASE Enablednofsgsbase30M60M90M120M150MSE +/- 833744.91, N = 3SE +/- 497372.75, N = 31422056481414481981. (CXX) g++ options: -O3 -march=native -std=c++11 -fno-builtin-memcmp -fno-rtti -rdynamic -lpthread

OpenBenchmarking.orgOp/s, More Is BetterFacebook RocksDB 6.3.6Test: Sequential FillFSGSBASE Enablednofsgsbase40K80K120K160K200KSE +/- 190.16, N = 3SE +/- 107.10, N = 31879501896431. (CXX) g++ options: -O3 -march=native -std=c++11 -fno-builtin-memcmp -fno-rtti -rdynamic -lpthread

OpenBenchmarking.orgOp/s, More Is BetterFacebook RocksDB 6.3.6Test: Random Fill SyncFSGSBASE Enablednofsgsbase14002800420056007000SE +/- 511.75, N = 15SE +/- 26.46, N = 3653256811. (CXX) g++ options: -O3 -march=native -std=c++11 -fno-builtin-memcmp -fno-rtti -rdynamic -lpthread

OpenBenchmarking.orgOp/s, More Is BetterFacebook RocksDB 6.3.6Test: Read While WritingFSGSBASE Enablednofsgsbase1.2M2.4M3.6M4.8M6MSE +/- 27153.16, N = 3SE +/- 54477.77, N = 3535627253964111. (CXX) g++ options: -O3 -march=native -std=c++11 -fno-builtin-memcmp -fno-rtti -rdynamic -lpthread

Flexible IO Tester

Fio is an advanced disk benchmark that depends upon the kernel's AIO access library. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgIOPS, More Is BetterFlexible IO Tester 3.18Type: Random Write - Engine: IO_uring - Buffered: Yes - Direct: No - Block Size: 2MB - Disk Target: Default Test DirectoryFSGSBASE Enablednofsgsbase4080120160200SE +/- 0.33, N = 3SE +/- 0.88, N = 31871351. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native -ll -lcurl -lssl -lcrypto -lnuma -libverbs -lrt -laio -lz -lpthread -lm -ldl

OpenBenchmarking.orgIOPS, More Is BetterFlexible IO Tester 3.18Type: Random Write - Engine: IO_uring - Buffered: Yes - Direct: No - Block Size: 4KB - Disk Target: Default Test DirectoryFSGSBASE Enablednofsgsbase20K40K60K80K100KSE +/- 251.66, N = 3SE +/- 100.00, N = 388800634001. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native -ll -lcurl -lssl -lcrypto -lnuma -libverbs -lrt -laio -lz -lpthread -lm -ldl

OpenBenchmarking.orgMB/s, More Is BetterFlexible IO Tester 3.18Type: Sequential Write - Engine: IO_uring - Buffered: Yes - Direct: No - Block Size: 2MB - Disk Target: Default Test DirectoryFSGSBASE Enablednofsgsbase80160240320400SE +/- 6.17, N = 3SE +/- 12.84, N = 153853461. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native -ll -lcurl -lssl -lcrypto -lnuma -libverbs -lrt -laio -lz -lpthread -lm -ldl

OpenBenchmarking.orgIOPS, More Is BetterFlexible IO Tester 3.18Type: Sequential Write - Engine: IO_uring - Buffered: Yes - Direct: No - Block Size: 2MB - Disk Target: Default Test DirectoryFSGSBASE Enablednofsgsbase4080120160200SE +/- 2.96, N = 3SE +/- 6.41, N = 151891701. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -fcommon -U_FORTIFY_SOURCE -march=native -ll -lcurl -lssl -lcrypto -lnuma -libverbs -lrt -laio -lz -lpthread -lm -ldl

GROMACS

The GROMACS (GROningen MAchine for Chemical Simulations) molecular dynamics package testing on the CPU with the water_GMX50 data. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgNs Per Day, More Is BetterGROMACS 2020.1Water BenchmarkFSGSBASE Enablednofsgsbase0.79071.58142.37213.16283.9535SE +/- 0.001, N = 3SE +/- 0.006, N = 33.5063.5141. (CXX) g++ options: -O3 -march=native -pthread -lrt -lpthread -lm

Himeno Benchmark

The Himeno benchmark is a linear solver of pressure Poisson using a point-Jacobi method. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMFLOPS, More Is BetterHimeno Benchmark 3.0Poisson Pressure SolverFSGSBASE Enablednofsgsbase9001800270036004500SE +/- 0.87, N = 3SE +/- 0.38, N = 33932.763966.361. (CC) gcc options: -O3 -march=native -mavx2

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: ReactorFSGSBASE Enablednofsgsbase60120180240300SE +/- 3.18, N = 3SE +/- 2.12, N = 3276.58288.79

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 5.3.1FSGSBASE Enablednofsgsbase90K180K270K360K450KSE +/- 4575.82, N = 3SE +/- 5862.96, N = 3428585.01418712.101. (CXX) g++ options: -O2 -levent -lpthread -lz -lpcre

LAMMPS Molecular Dynamics Simulator

LAMMPS is a classical molecular dynamics code, and an acronym for Large-scale Atomic/Molecular Massively Parallel Simulator. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgns/day, More Is BetterLAMMPS Molecular Dynamics Simulator 9Jan2020Model: Rhodopsin ProteinFSGSBASE Enablednofsgsbase48121620SE +/- 0.30, N = 3SE +/- 0.31, N = 1517.9517.141. (CXX) g++ options: -O3 -march=native -rdynamic -ljpeg -lpng -lz -lfftw3 -lm

LevelDB

LevelDB is a key-value storage library developed by Google that supports making use of Snappy for data compression and has other modern features. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMicroseconds Per Op, Fewer Is BetterLevelDB 1.22Benchmark: Hot ReadFSGSBASE Enablednofsgsbase20406080100SE +/- 1.10, N = 3SE +/- 1.38, N = 392.2891.651. (CXX) g++ options: -O3 -march=native -lsnappy -lpthread

OpenBenchmarking.orgMB/s, More Is BetterLevelDB 1.22Benchmark: Fill SyncFSGSBASE Enablednofsgsbase0.4050.811.2151.622.025SE +/- 0.00, N = 3SE +/- 0.00, N = 31.81.81. (CXX) g++ options: -O3 -march=native -lsnappy -lpthread

OpenBenchmarking.orgMicroseconds Per Op, Fewer Is BetterLevelDB 1.22Benchmark: Fill SyncFSGSBASE Enablednofsgsbase10002000300040005000SE +/- 8.32, N = 3SE +/- 8.46, N = 34485.354460.311. (CXX) g++ options: -O3 -march=native -lsnappy -lpthread

OpenBenchmarking.orgMB/s, More Is BetterLevelDB 1.22Benchmark: OverwriteFSGSBASE Enablednofsgsbase3691215SE +/- 0.03, N = 3SE +/- 0.03, N = 39.910.01. (CXX) g++ options: -O3 -march=native -lsnappy -lpthread

OpenBenchmarking.orgMicroseconds Per Op, Fewer Is BetterLevelDB 1.22Benchmark: OverwriteFSGSBASE Enablednofsgsbase2004006008001000SE +/- 1.81, N = 3SE +/- 2.31, N = 3799.36792.441. (CXX) g++ options: -O3 -march=native -lsnappy -lpthread

OpenBenchmarking.orgMB/s, More Is BetterLevelDB 1.22Benchmark: Random FillFSGSBASE Enablednofsgsbase3691215SE +/- 0.03, N = 3SE +/- 0.10, N = 39.810.01. (CXX) g++ options: -O3 -march=native -lsnappy -lpthread

OpenBenchmarking.orgMicroseconds Per Op, Fewer Is BetterLevelDB 1.22Benchmark: Random FillFSGSBASE Enablednofsgsbase2004006008001000SE +/- 3.59, N = 3SE +/- 8.40, N = 3809.70798.601. (CXX) g++ options: -O3 -march=native -lsnappy -lpthread

OpenBenchmarking.orgMicroseconds Per Op, Fewer Is BetterLevelDB 1.22Benchmark: Random ReadFSGSBASE Enablednofsgsbase20406080100SE +/- 0.05, N = 3SE +/- 0.66, N = 394.0093.011. (CXX) g++ options: -O3 -march=native -lsnappy -lpthread

OpenBenchmarking.orgMicroseconds Per Op, Fewer Is BetterLevelDB 1.22Benchmark: Seek RandomFSGSBASE Enablednofsgsbase306090120150SE +/- 0.13, N = 3SE +/- 0.75, N = 3113.58113.201. (CXX) g++ options: -O3 -march=native -lsnappy -lpthread

OpenBenchmarking.orgMicroseconds Per Op, Fewer Is BetterLevelDB 1.22Benchmark: Random DeleteFSGSBASE Enablednofsgsbase170340510680850SE +/- 1.02, N = 3SE +/- 2.15, N = 3774.36761.431. (CXX) g++ options: -O3 -march=native -lsnappy -lpthread

OpenBenchmarking.orgMB/s, More Is BetterLevelDB 1.22Benchmark: Sequential FillFSGSBASE Enablednofsgsbase3691215SE +/- 0.03, N = 3SE +/- 0.03, N = 39.69.91. (CXX) g++ options: -O3 -march=native -lsnappy -lpthread

OpenBenchmarking.orgMicroseconds Per Op, Fewer Is BetterLevelDB 1.22Benchmark: Sequential FillFSGSBASE Enablednofsgsbase2004006008001000SE +/- 3.79, N = 3SE +/- 3.18, N = 3826.27809.441. (CXX) g++ options: -O3 -march=native -lsnappy -lpthread

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 10.5.2Clients: 64FSGSBASE Enablednofsgsbase4080120160200SE +/- 2.42, N = 6SE +/- 2.50, N = 52051991. (CXX) g++ options: -O3 -march=native -pie -fPIC -fstack-protector -O2 -lpthread -llzma -lbz2 -lsnappy -laio -lnuma -lpcre2-8 -lcrypt -lz -lm -lssl -lcrypto -ldl

OpenBenchmarking.orgQueries Per Second, More Is BetterMariaDB 10.5.2Clients: 128FSGSBASE Enablednofsgsbase4080120160200SE +/- 0.48, N = 3SE +/- 0.65, N = 31591541. (CXX) g++ options: -O3 -march=native -pie -fPIC -fstack-protector -O2 -lpthread -llzma -lbz2 -lsnappy -laio -lnuma -lpcre2-8 -lcrypt -lz -lm -lssl -lcrypto -ldl

OpenBenchmarking.orgQueries Per Second, More Is BetterMariaDB 10.5.2Clients: 256FSGSBASE Enablednofsgsbase306090120150SE +/- 0.58, N = 3SE +/- 0.32, N = 31501431. (CXX) g++ options: -O3 -march=native -pie -fPIC -fstack-protector -O2 -lpthread -llzma -lbz2 -lsnappy -laio -lnuma -lpcre2-8 -lcrypt -lz -lm -lssl -lcrypto -ldl

OpenBenchmarking.orgQueries Per Second, More Is BetterMariaDB 10.5.2Clients: 512FSGSBASE Enablednofsgsbase4080120160200SE +/- 2.84, N = 9SE +/- 0.52, N = 31611401. (CXX) g++ options: -O3 -march=native -pie -fPIC -fstack-protector -O2 -lpthread -llzma -lbz2 -lsnappy -laio -lnuma -lpcre2-8 -lcrypt -lz -lm -lssl -lcrypto -ldl

Memtier_benchmark

Memtier_benchmark is a NoSQL Redis/Memcache traffic generation plus benchmarking tool. This current test profile currently just stresses the Redis protocol and basic options exposed wotj a 1:1 Set/Get ratio, 30 pipeline, 100 clients per thread, and thread count equal to the number of CPU cores/threads present. Patches to extend the test are welcome as always. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgOps/sec, More Is BetterMemtier_benchmark 1.2.17Protocol: RedisFSGSBASE Enablednofsgsbase600K1200K1800K2400K3000KSE +/- 74458.90, N = 15SE +/- 13554.01, N = 32859981.402635763.821. (CXX) g++ options: -O2 -levent -lpthread -lz -lpcre

Mlpack Benchmark

Mlpack benchmark scripts for machine learning libraries Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterMlpack BenchmarkBenchmark: scikit_icaFSGSBASE Enablednofsgsbase20406080100SE +/- 0.14, N = 3SE +/- 0.44, N = 374.6675.12

OpenBenchmarking.orgSeconds, Fewer Is BetterMlpack BenchmarkBenchmark: scikit_qdaFSGSBASE Enablednofsgsbase1020304050SE +/- 0.77, N = 12SE +/- 0.45, N = 1146.0346.09

OpenBenchmarking.orgSeconds, Fewer Is BetterMlpack BenchmarkBenchmark: scikit_svmFSGSBASE Enablednofsgsbase714212835SE +/- 0.14, N = 3SE +/- 0.02, N = 328.7128.46

OpenBenchmarking.orgSeconds, Fewer Is BetterMlpack BenchmarkBenchmark: scikit_linearridgeregressionFSGSBASE Enablednofsgsbase0.45230.90461.35691.80922.2615SE +/- 0.02, N = 9SE +/- 0.03, N = 151.962.01

NAMD

NAMD is a parallel molecular dynamics code designed for high-performance simulation of large biomolecular systems. NAMD was developed by the Theoretical and Computational Biophysics Group in the Beckman Institute for Advanced Science and Technology at the University of Illinois at Urbana-Champaign. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgdays/ns, Fewer Is BetterNAMD 2.13ATPase Simulation - 327,506 AtomsFSGSBASE Enablednofsgsbase0.13750.2750.41250.550.6875SE +/- 0.00455, N = 14SE +/- 0.00071, N = 30.611040.61077

Node.js Express HTTP Load Test

A Node.js Express server with a Node-based loadtest client for facilitating HTTP benchmarking. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgRequests Per Second, More Is BetterNode.js Express HTTP Load TestFSGSBASE Enablednofsgsbase2K4K6K8K10KSE +/- 152.70, N = 15SE +/- 181.89, N = 15939593421. Nodejs v10.19.0

Numenta Anomaly Benchmark

Numenta Anomaly Benchmark (NAB) is a benchmark for evaluating algorithms for anomaly detection in streaming, real-time applications. It is comprised of over 50 labeled real-world and artificial timeseries data files plus a novel scoring mechanism designed for real-time applications. This test profile currently measures the time to run various detectors. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterNumenta Anomaly Benchmark 1.1Detector: EXPoSEFSGSBASE Enablednofsgsbase30060090012001500SE +/- 15.42, N = 3SE +/- 5.29, N = 31513.781500.85

OpenBenchmarking.orgSeconds, Fewer Is BetterNumenta Anomaly Benchmark 1.1Detector: Relative EntropyFSGSBASE Enablednofsgsbase48121620SE +/- 0.05, N = 3SE +/- 0.16, N = 314.8114.91

OpenBenchmarking.orgSeconds, Fewer Is BetterNumenta Anomaly Benchmark 1.1Detector: Earthgecko SkylineFSGSBASE Enablednofsgsbase20406080100SE +/- 0.50, N = 3SE +/- 0.29, N = 390.7889.61

OpenBenchmarking.orgSeconds, Fewer Is BetterNumenta Anomaly Benchmark 1.1Detector: Bayesian ChangepointFSGSBASE Enablednofsgsbase816243240SE +/- 0.35, N = 3SE +/- 0.26, N = 332.7432.60

oneDNN

This is a test of the Intel oneDNN as an Intel-optimized library for Deep Neural Networks and making use of its built-in benchdnn functionality. The result is the total perf time reported. Intel oneDNN was formerly known as DNNL (Deep Neural Network Library) and MKL-DNN before being rebranded as part of the oneAPI initiative. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 1.5Harness: IP Batch 1D - Data Type: bf16bf16bf16 - Engine: CPUFSGSBASE Enablednofsgsbase1.28152.5633.84455.1266.4075SE +/- 0.00089, N = 3SE +/- 0.00720, N = 35.695625.67910MIN: 5.52MIN: 5.51. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 1.5Harness: IP Batch All - Data Type: bf16bf16bf16 - Engine: CPUFSGSBASE Enablednofsgsbase1224364860SE +/- 0.02, N = 3SE +/- 0.03, N = 351.1051.11MIN: 50.05MIN: 50.211. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 1.5Harness: Convolution Batch Shapes Auto - Data Type: bf16bf16bf16 - Engine: CPUFSGSBASE Enablednofsgsbase246810SE +/- 0.00089, N = 3SE +/- 0.01144, N = 36.397286.38735MIN: 6.3MIN: 6.31. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 1.5Harness: Deconvolution Batch deconv_1d - Data Type: bf16bf16bf16 - Engine: CPUFSGSBASE Enablednofsgsbase246810SE +/- 0.01137, N = 3SE +/- 0.00338, N = 37.391547.39063MIN: 7.23MIN: 7.231. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 1.5Harness: Deconvolution Batch deconv_3d - Data Type: bf16bf16bf16 - Engine: CPUFSGSBASE Enablednofsgsbase3691215SE +/- 0.00175, N = 3SE +/- 0.00883, N = 39.461589.46163MIN: 9.31MIN: 9.351. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

OpenBenchmarking.orgms, Fewer Is BetteroneDNN 1.5Harness: Matrix Multiply Batch Shapes Transformer - Data Type: bf16bf16bf16 - Engine: CPUFSGSBASE Enablednofsgsbase0.32670.65340.98011.30681.6335SE +/- 0.00269, N = 3SE +/- 0.00164, N = 31.449231.45193MIN: 1.4MIN: 1.411. (CXX) g++ options: -O3 -march=native -std=c++11 -fopenmp -msse4.1 -fPIC -pie -lpthread -ldl

PlaidML

This test profile uses PlaidML deep learning framework developed by Intel for offering up various benchmarks. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFPS, More Is BetterPlaidMLFP16: No - Mode: Inference - Network: VGG16 - Device: CPUFSGSBASE Enablednofsgsbase612182430SE +/- 0.24, N = 3SE +/- 0.07, N = 324.9825.15

OpenBenchmarking.orgFPS, More Is BetterPlaidMLFP16: No - Mode: Inference - Network: VGG19 - Device: CPUFSGSBASE Enablednofsgsbase510152025SE +/- 0.18, N = 3SE +/- 0.12, N = 320.6421.39

OpenBenchmarking.orgFPS, More Is BetterPlaidMLFP16: No - Mode: Inference - Network: IMDB LSTM - Device: CPUFSGSBASE Enablednofsgsbase2004006008001000SE +/- 11.90, N = 4SE +/- 3.24, N = 3868.09850.65

OpenBenchmarking.orgFPS, More Is BetterPlaidMLFP16: No - Mode: Inference - Network: Mobilenet - Device: CPUFSGSBASE Enablednofsgsbase3691215SE +/- 0.12, N = 3SE +/- 0.09, N = 310.6210.60

OpenBenchmarking.orgFPS, More Is BetterPlaidMLFP16: No - Mode: Inference - Network: ResNet 50 - Device: CPUFSGSBASE Enablednofsgsbase1.0082.0163.0244.0325.04SE +/- 0.03, N = 3SE +/- 0.03, N = 34.454.48

OpenBenchmarking.orgFPS, More Is BetterPlaidMLFP16: No - Mode: Inference - Network: DenseNet 201 - Device: CPUFSGSBASE Enablednofsgsbase0.44550.8911.33651.7822.2275SE +/- 0.01, N = 3SE +/- 0.01, N = 31.961.98

OpenBenchmarking.orgFPS, More Is BetterPlaidMLFP16: No - Mode: Inference - Network: Inception V3 - Device: CPUFSGSBASE Enablednofsgsbase1.23752.4753.71254.956.1875SE +/- 0.03, N = 3SE +/- 0.02, N = 35.505.45

OpenBenchmarking.orgFPS, More Is BetterPlaidMLFP16: No - Mode: Inference - Network: NASNer Large - Device: CPUFSGSBASE Enablednofsgsbase0.13050.2610.39150.5220.6525SE +/- 0.00, N = 3SE +/- 0.00, N = 30.570.58

pmbench

Pmbench is a Linux paging and virtual memory benchmark. This test profile will report the average page latency of the system. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgus - Average Page Latency, Fewer Is BetterpmbenchConcurrent Worker Threads: 72 - Read-Write Ratio: 100% ReadsFSGSBASE Enablednofsgsbase0.01040.02080.03120.04160.052SE +/- 0.0012, N = 12SE +/- 0.0004, N = 150.04600.04511. (CC) gcc options: -lm -luuid -lxml2 -m64 -pthread

OpenBenchmarking.orgus - Average Page Latency, Fewer Is BetterpmbenchConcurrent Worker Threads: 72 - Read-Write Ratio: 100% WritesFSGSBASE Enablednofsgsbase0.01830.03660.05490.07320.0915SE +/- 0.0010, N = 5SE +/- 0.0009, N = 30.08020.08121. (CC) gcc options: -lm -luuid -lxml2 -m64 -pthread

OpenBenchmarking.orgus - Average Page Latency, Fewer Is BetterpmbenchConcurrent Worker Threads: 1 - Read-Write Ratio: 80% Reads 20% WritesFSGSBASE Enablednofsgsbase0.0170.0340.0510.0680.085SE +/- 0.0002, N = 3SE +/- 0.0002, N = 30.07560.07561. (CC) gcc options: -lm -luuid -lxml2 -m64 -pthread

PostgreSQL pgbench

This is a simple benchmark of PostgreSQL using pgbench. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL pgbench 12.0Scaling: Buffer Test - Test: Normal Load - Mode: Read OnlyFSGSBASE Enablednofsgsbase130K260K390K520K650KSE +/- 2139.37, N = 3SE +/- 3396.56, N = 3593329.43594395.911. (CC) gcc options: -fno-strict-aliasing -fwrapv -O3 -march=native -lpgcommon -lpgport -lpq -lpthread -lrt -lcrypt -ldl -lm

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL pgbench 12.0Scaling: Buffer Test - Test: Normal Load - Mode: Read WriteFSGSBASE Enablednofsgsbase11002200330044005500SE +/- 59.24, N = 6SE +/- 34.32, N = 34908.502762.901. (CC) gcc options: -fno-strict-aliasing -fwrapv -O3 -march=native -lpgcommon -lpgport -lpq -lpthread -lrt -lcrypt -ldl -lm

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL pgbench 12.0Scaling: Buffer Test - Test: Heavy Contention - Mode: Read OnlyFSGSBASE Enablednofsgsbase130K260K390K520K650KSE +/- 1415.39, N = 3SE +/- 680.20, N = 3618994.74619694.861. (CC) gcc options: -fno-strict-aliasing -fwrapv -O3 -march=native -lpgcommon -lpgport -lpq -lpthread -lrt -lcrypt -ldl -lm

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL pgbench 12.0Scaling: Buffer Test - Test: Heavy Contention - Mode: Read WriteFSGSBASE Enablednofsgsbase10002000300040005000SE +/- 59.91, N = 3SE +/- 25.55, N = 94727.712634.271. (CC) gcc options: -fno-strict-aliasing -fwrapv -O3 -march=native -lpgcommon -lpgport -lpq -lpthread -lrt -lcrypt -ldl -lm

PostMark

This is a test of NetApp's PostMark benchmark designed to simulate small-file testing similar to the tasks endured by web and mail servers. This test profile will set PostMark to perform 25,000 transactions with 500 files simultaneously with the file sizes ranging between 5 and 512 kilobytes. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgTPS, More Is BetterPostMark 1.51Disk Transaction PerformanceFSGSBASE Enablednofsgsbase12002400360048006000SE +/- 44.00, N = 3SE +/- 44.00, N = 3572557251. (CC) gcc options: -O3 -march=native

QMCPACK

QMCPACK is a modern high-performance open-source Quantum Monte Carlo (QMC) simulation code making use of MPI for this benchmark of the H20 example code. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgTotal Execution Time - Seconds, Fewer Is BetterQMCPACK 3.8FSGSBASE Enablednofsgsbase60012001800240030002687.62688.51. (CXX) g++ options: -O3 -march=native -fopenmp -fomit-frame-pointer -finline-limit=1000 -fstrict-aliasing -funroll-all-loops -ffast-math -lm

Redis

Redis is an open-source data structure server. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgRequests Per Second, More Is BetterRedis 5.0.5Test: LPOPFSGSBASE Enablednofsgsbase600K1200K1800K2400K3000KSE +/- 14141.11, N = 3SE +/- 9534.81, N = 32611119.751707562.791. (CXX) g++ options: -MM -MT -g3 -fvisibility=hidden -O3 -march=native

OpenBenchmarking.orgRequests Per Second, More Is BetterRedis 5.0.5Test: SADDFSGSBASE Enablednofsgsbase400K800K1200K1600K2000KSE +/- 2516.35, N = 3SE +/- 30688.88, N = 152087688.752071005.921. (CXX) g++ options: -MM -MT -g3 -fvisibility=hidden -O3 -march=native

OpenBenchmarking.orgRequests Per Second, More Is BetterRedis 5.0.5Test: GETFSGSBASE Enablednofsgsbase500K1000K1500K2000K2500KSE +/- 31845.75, N = 3SE +/- 18114.33, N = 32500801.502340376.201. (CXX) g++ options: -MM -MT -g3 -fvisibility=hidden -O3 -march=native

OpenBenchmarking.orgRequests Per Second, More Is BetterRedis 5.0.5Test: SETFSGSBASE Enablednofsgsbase400K800K1200K1600K2000KSE +/- 2456.08, N = 3SE +/- 4205.47, N = 31918164.831908415.461. (CXX) g++ options: -MM -MT -g3 -fvisibility=hidden -O3 -march=native

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 0.10.0Test: Apache Spark ALSFSGSBASE Enablednofsgsbase6001200180024003000SE +/- 32.22, N = 25SE +/- 41.50, N = 252896.192985.60

OpenBenchmarking.orgms, Fewer Is BetterRenaissance 0.10.0Test: Savina Reactors.IOFSGSBASE Enablednofsgsbase5K10K15K20K25KSE +/- 370.01, N = 20SE +/- 803.93, N = 1621837.0024407.08

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 0.11.07Test: AtomicFSGSBASE Enablednofsgsbase20K40K60K80K100KSE +/- 1267.27, N = 3SE +/- 1320.48, N = 390016.1885800.801. (CC) gcc options: -O3 -march=native -O2 -std=gnu99 -lm -laio -lcrypt -lrt -lz -ldl -lpthread -lc

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.11.07Test: SENDFILEFSGSBASE Enablednofsgsbase100K200K300K400K500KSE +/- 1247.48, N = 3SE +/- 104.27, N = 3447930.47444432.101. (CC) gcc options: -O3 -march=native -O2 -std=gnu99 -lm -laio -lcrypt -lrt -lz -ldl -lpthread -lc

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.11.07Test: CPU StressFSGSBASE Enablednofsgsbase3K6K9K12K15KSE +/- 19.86, N = 3SE +/- 52.61, N = 311896.4211983.661. (CC) gcc options: -O3 -march=native -O2 -std=gnu99 -lm -laio -lcrypt -lrt -lz -ldl -lpthread -lc

OpenBenchmarking.orgBogo Ops/s, More Is BetterStress-NG 0.11.07Test: Context SwitchingFSGSBASE Enablednofsgsbase2M4M6M8M10MSE +/- 154150.18, N = 3SE +/- 27584.57, N = 39410762.677847877.591. (CC) gcc options: -O3 -march=native -O2 -std=gnu99 -lm -laio -lcrypt -lrt -lz -ldl -lpthread -lc

SVT-AV1

This is a test of the Intel Open Visual Cloud Scalable Video Technology SVT-AV1 CPU-based multi-threaded video encoder for the AV1 video format with a sample 1080p YUV video file. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 0.8Encoder Mode: Enc Mode 0 - Input: 1080pFSGSBASE Enablednofsgsbase0.0270.0540.0810.1080.135SE +/- 0.000, N = 3SE +/- 0.000, N = 30.1200.1201. (CXX) g++ options: -O3 -fcommon -fPIE -fPIC -pie

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 0.8Encoder Mode: Enc Mode 4 - Input: 1080pFSGSBASE Enablednofsgsbase1.29832.59663.89495.19326.4915SE +/- 0.067, N = 3SE +/- 0.082, N = 35.7705.6831. (CXX) g++ options: -O3 -fcommon -fPIE -fPIC -pie

OpenBenchmarking.orgFrames Per Second, More Is BetterSVT-AV1 0.8Encoder Mode: Enc Mode 8 - Input: 1080pFSGSBASE Enablednofsgsbase1122334455SE +/- 0.53, N = 3SE +/- 0.04, N = 348.8749.211. (CXX) g++ options: -O3 -fcommon -fPIE -fPIC -pie

Timed Apache Compilation

This test times how long it takes to build the Apache HTTPD web server. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Apache Compilation 2.4.41Time To CompileFSGSBASE Enablednofsgsbase612182430SE +/- 0.03, N = 3SE +/- 0.03, N = 325.7625.47

Timed GDB GNU Debugger Compilation

This test times how long it takes to build the GNU Debugger (GDB) in a default configuration. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed GDB GNU Debugger Compilation 9.1Time To CompileFSGSBASE Enablednofsgsbase306090120150SE +/- 0.07, N = 3SE +/- 0.05, N = 3137.40135.38

Timed HMMer Search

This test searches through the Pfam database of profile hidden markov models. The search finds the domain structure of Drosophila Sevenless protein. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed HMMer Search 2.3.2Pfam Database SearchFSGSBASE Enablednofsgsbase3691215SE +/- 0.17, N = 3SE +/- 0.10, N = 312.7212.731. (CC) gcc options: -O3 -march=native -pthread -lhmmer -lsquid -lm

Timed Linux Kernel Compilation

This test times how long it takes to build the Linux kernel in a default configuration. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed Linux Kernel Compilation 5.4Time To CompileFSGSBASE Enablednofsgsbase918273645SE +/- 0.38, N = 8SE +/- 0.39, N = 837.4337.18

Timed LLVM Compilation

This test times how long it takes to build the LLVM compiler. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed LLVM Compilation 10.0Time To CompileFSGSBASE Enablednofsgsbase60120180240300SE +/- 3.74, N = 4SE +/- 1.49, N = 3283.74285.30

Timed MAFFT Alignment

This test performs an alignment of 100 pyruvate decarboxylase sequences. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed MAFFT Alignment 7.392Multiple Sequence AlignmentFSGSBASE Enablednofsgsbase0.59961.19921.79882.39842.998SE +/- 0.054, N = 15SE +/- 0.056, N = 152.6602.6651. (CC) gcc options: -std=c99 -O3 -lm -lpthread

VP9 libvpx Encoding

This is a standard video encoding performance test of Google's libvpx library and the vpxenc command for the VP9/WebM format using a sample 1080p video. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFrames Per Second, More Is BetterVP9 libvpx Encoding 1.8.2Speed: Speed 0FSGSBASE Enablednofsgsbase246810SE +/- 0.01, N = 3SE +/- 0.02, N = 36.146.121. (CXX) g++ options: -m64 -lm -lpthread -O3 -march=native -fPIC -U_FORTIFY_SOURCE -std=c++11

OpenBenchmarking.orgFrames Per Second, More Is BetterVP9 libvpx Encoding 1.8.2Speed: Speed 5FSGSBASE Enablednofsgsbase612182430SE +/- 0.09, N = 3SE +/- 0.08, N = 323.0123.241. (CXX) g++ options: -m64 -lm -lpthread -O3 -march=native -fPIC -U_FORTIFY_SOURCE -std=c++11

YafaRay

YafaRay is an open-source physically based montecarlo ray-tracing engine. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterYafaRay 3.4.1Total Time For Sample SceneFSGSBASE Enablednofsgsbase306090120150SE +/- 2.93, N = 15SE +/- 3.44, N = 15108.82113.631. (CXX) g++ options: -std=c++11 -O3 -ffast-math -rdynamic -ldl -lImath -lIlmImf -lIex -lHalf -lz -lIlmThread -lxml2 -lfreetype -lpthread

111 Results Shown

AOM AV1:
  Speed 0 Two-Pass
  Speed 4 Two-Pass
  Speed 6 Realtime
  Speed 6 Two-Pass
  Speed 8 Realtime
Apache Cassandra
Apache HBase:
  Increment - 1:
    Rows Per Second
    Microseconds - Average Latency
  Rand Read - 1:
    Rows Per Second
    Microseconds - Average Latency
  Seq Read - 1:
    Rows Per Second
    Microseconds - Average Latency
  Async Rand Read - 1:
    Rows Per Second
    Microseconds - Average Latency
Apache Siege:
  10
  50
  200
BlogBench
CP2K Molecular Dynamics
ctx_clock
dav1d:
  Chimera 1080p
  Summer Nature 4K
  Summer Nature 1080p
  Chimera 1080p 10-bit
ebizzy
Facebook RocksDB:
  Rand Fill
  Rand Read
  Seq Fill
  Rand Fill Sync
  Read While Writing
Flexible IO Tester:
  Rand Write - IO_uring - Yes - No - 2MB - Default Test Directory
  Rand Write - IO_uring - Yes - No - 4KB - Default Test Directory
  Seq Write - IO_uring - Yes - No - 2MB - Default Test Directory
  Seq Write - IO_uring - Yes - No - 2MB - Default Test Directory
GROMACS
Himeno Benchmark
Java Gradle Build
KeyDB
LAMMPS Molecular Dynamics Simulator
LevelDB:
  Hot Read
  Fill Sync
  Fill Sync
  Overwrite
  Overwrite
  Rand Fill
  Rand Fill
  Rand Read
  Seek Rand
  Rand Delete
  Seq Fill
  Seq Fill
MariaDB:
  64
  128
  256
  512
Memtier_benchmark
Mlpack Benchmark:
  scikit_ica
  scikit_qda
  scikit_svm
  scikit_linearridgeregression
NAMD
Node.js Express HTTP Load Test
Numenta Anomaly Benchmark:
  EXPoSE
  Relative Entropy
  Earthgecko Skyline
  Bayesian Changepoint
oneDNN:
  IP Batch 1D - bf16bf16bf16 - CPU
  IP Batch All - bf16bf16bf16 - CPU
  Convolution Batch Shapes Auto - bf16bf16bf16 - CPU
  Deconvolution Batch deconv_1d - bf16bf16bf16 - CPU
  Deconvolution Batch deconv_3d - bf16bf16bf16 - CPU
  Matrix Multiply Batch Shapes Transformer - bf16bf16bf16 - CPU
PlaidML:
  No - Inference - VGG16 - CPU
  No - Inference - VGG19 - CPU
  No - Inference - IMDB LSTM - CPU
  No - Inference - Mobilenet - CPU
  No - Inference - ResNet 50 - CPU
  No - Inference - DenseNet 201 - CPU
  No - Inference - Inception V3 - CPU
  No - Inference - NASNer Large - CPU
pmbench:
  72 - 100% Reads
  72 - 100% Writes
  1 - 80% Reads 20% Writes
PostgreSQL pgbench:
  Buffer Test - Normal Load - Read Only
  Buffer Test - Normal Load - Read Write
  Buffer Test - Heavy Contention - Read Only
  Buffer Test - Heavy Contention - Read Write
PostMark
QMCPACK
Redis:
  LPOP
  SADD
  GET
  SET
Renaissance:
  Apache Spark ALS
  Savina Reactors.IO
Stress-NG:
  Atomic
  SENDFILE
  CPU Stress
  Context Switching
SVT-AV1:
  Enc Mode 0 - 1080p
  Enc Mode 4 - 1080p
  Enc Mode 8 - 1080p
Timed Apache Compilation
Timed GDB GNU Debugger Compilation
Timed HMMer Search
Timed Linux Kernel Compilation
Timed LLVM Compilation
Timed MAFFT Alignment
VP9 libvpx Encoding:
  Speed 0
  Speed 5
YafaRay

FSGSBASE Enabled

Environment Notes: CXXFLAGS="-O3 -march=native" CFLAGS="-O3 -march=native"
Compiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --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++,gm2 --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none,hsa --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-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v
Disk Notes: MQ-DEADLINE / errors=remount-ro,relatime,rw
Processor Notes: Scaling Governor: intel_pstate powersave - CPU Microcode: 0x500002c
Java Notes: OpenJDK Runtime Environment (build 11.0.7-ea+9-post-Ubuntu-1ubuntu1)
Python Notes: Python 3.8.2
Security Notes: itlb_multihit: KVM: Mitigation of Split huge pages + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced IBRS IBPB: conditional RSB filling + srbds: Not affected + tsx_async_abort: Mitigation of TSX disabled

Testing initiated at 20 June 2020 20:34 by user phoronix.

nofsgsbase

Processor: 2 x Intel Xeon Gold 5220R @ 3.90GHz (36 Cores / 72 Threads), Motherboard: TYAN S7106 (V2.01.B40 BIOS), Chipset: Intel Sky Lake-E DMI3 Registers, Memory: 94GB, Disk: 500GB Samsung SSD 860, Graphics: ASPEED, Monitor: VE228, Network: 2 x Intel I210 + 2 x QLogic cLOM8214 1/10GbE

OS: Ubuntu 20.04, Kernel: 5.8.0-rc1-phx-fsgsbase (x86_64) 20200620, Desktop: GNOME Shell 3.36.1, Display Server: X Server 1.20.8, Display Driver: modesetting 1.20.8, Compiler: GCC 9.3.0, File-System: ext4, Screen Resolution: 1920x1080

Environment Notes: CXXFLAGS="-O3 -march=native" CFLAGS="-O3 -march=native"
Compiler Notes: --build=x86_64-linux-gnu --disable-vtable-verify --disable-werror --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++,gm2 --enable-libstdcxx-debug --enable-libstdcxx-time=yes --enable-multiarch --enable-multilib --enable-nls --enable-objc-gc=auto --enable-offload-targets=nvptx-none,hsa --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-default-libstdcxx-abi=new --with-gcc-major-version-only --with-multilib-list=m32,m64,mx32 --with-target-system-zlib=auto --with-tune=generic --without-cuda-driver -v
Disk Notes: MQ-DEADLINE / errors=remount-ro,relatime,rw
Processor Notes: Scaling Governor: intel_pstate powersave - CPU Microcode: 0x5002f01
Java Notes: OpenJDK Runtime Environment (build 11.0.7+10-post-Ubuntu-3ubuntu1)
Python Notes: Python 3.8.2
Security Notes: itlb_multihit: KVM: Mitigation of Split huge pages + l1tf: Not affected + mds: Not affected + meltdown: Not affected + spec_store_bypass: Mitigation of SSB disabled via prctl and seccomp + spectre_v1: Mitigation of usercopy/swapgs barriers and __user pointer sanitization + spectre_v2: Mitigation of Enhanced IBRS IBPB: conditional RSB filling + srbds: Not affected + tsx_async_abort: Mitigation of TSX disabled

Testing initiated at 22 June 2020 16:05 by user phoronix.