Linux 5.6 IO Scheduler Benchmarks

Linux 5.6 I/O scheduler benchmarks on NVMe SSD. 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 2003265-NI-LINUX56IO44
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

Limit displaying results to tests within:

Timed Code Compilation 2 Tests
C/C++ Compiler Tests 4 Tests
CPU Massive 3 Tests
Database Test Suite 8 Tests
Disk Test Suite 3 Tests
Common Kernel Benchmarks 3 Tests
Multi-Core 5 Tests
Programmer / Developer System Benchmarks 2 Tests
Server 8 Tests
Server CPU Tests 2 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
None
March 23 2020
  4 Hours
MQ-Deadline
March 23 2020
  8 Hours, 19 Minutes
Kyber
March 24 2020
  6 Hours, 57 Minutes
BFQ low-latency
March 24 2020
  8 Hours, 23 Minutes
BFQ
March 25 2020
  6 Hours, 51 Minutes
Invert Hiding All Results Option
  6 Hours, 54 Minutes

Only show results where is faster than
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):

None

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: NONE / errors=remount-ro,relatime,rw
Processor Notes: Scaling Governor: acpi-cpufreq ondemand - CPU Microcode: 0x8301025
Java Notes: OpenJDK Runtime Environment (build 11.0.6+10-post-Ubuntu-2ubuntu2)
Security Notes: itlb_multihit: Not affected + 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 Full AMD retpoline IBPB: conditional STIBP: conditional RSB filling + tsx_async_abort: Not affected

MQ-Deadline

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: acpi-cpufreq ondemand - CPU Microcode: 0x8301025
Java Notes: OpenJDK Runtime Environment (build 11.0.6+10-post-Ubuntu-2ubuntu2)
Security Notes: itlb_multihit: Not affected + 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 Full AMD retpoline IBPB: conditional STIBP: conditional RSB filling + tsx_async_abort: Not affected

Kyber

Processor: AMD Ryzen Threadripper 3970X 32-Core @ 3.70GHz (32 Cores / 64 Threads), Motherboard: ASUS ROG ZENITH II EXTREME (0702 BIOS), Chipset: AMD Starship/Matisse, Memory: 64GB, Disk: 1000GB Force MP600, Graphics: Gigabyte AMD Radeon RX 5500/5500M / Pro 5500M 8GB (1890/875MHz), Audio: AMD Navi 10 HDMI Audio, Monitor: ASUS MG28U, Network: Aquantia AQC107 NBase-T/IEEE + Intel I211 + Intel Wi-Fi 6 AX200

OS: Ubuntu 20.04, Kernel: 5.6.0-050600rc7-generic (x86_64) 20200322, Desktop: GNOME Shell 3.35.91, Display Server: X Server 1.20.7, Display Driver: modesetting 1.20.7, OpenGL: 4.6 Mesa 20.0.0 (LLVM 9.0.1), Compiler: GCC 9.3.0, File-System: ext4, Screen Resolution: 1280x800

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: KYBER / errors=remount-ro,relatime,rw
Processor Notes: Scaling Governor: acpi-cpufreq ondemand - CPU Microcode: 0x8301025
Java Notes: OpenJDK Runtime Environment (build 11.0.6+10-post-Ubuntu-2ubuntu2)
Security Notes: itlb_multihit: Not affected + 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 Full AMD retpoline IBPB: conditional STIBP: conditional RSB filling + tsx_async_abort: Not affected

BFQ low-latency

OS: Ubuntu 20.04, Kernel: 5.6.0-050600rc7-generic (x86_64) 20200322, Desktop: GNOME Shell 3.35.91, Display Server: X Server 1.20.7, Display Driver: modesetting 1.20.7, OpenGL: 4.6 Mesa 20.0.0 (LLVM 9.0.1), Compiler: GCC 9.3.0, File-System: ext4, Screen Resolution: 3840x2160

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: BFQ / errors=remount-ro,relatime,rw
Processor Notes: Scaling Governor: acpi-cpufreq ondemand - CPU Microcode: 0x8301025
Java Notes: OpenJDK Runtime Environment (build 11.0.6+10-post-Ubuntu-2ubuntu2)
Security Notes: itlb_multihit: Not affected + 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 Full AMD retpoline IBPB: conditional STIBP: conditional RSB filling + tsx_async_abort: Not affected

BFQ

OS: Ubuntu 20.04, Kernel: 5.6.0-050600rc7-generic (x86_64) 20200322, Desktop: GNOME Shell 3.35.91, Display Server: X Server 1.20.7, Display Driver: modesetting 1.20.7, OpenGL: 4.6 Mesa 20.0.0 (LLVM 9.0.1), Compiler: GCC 9.3.0, File-System: ext4, Screen Resolution: 1280x800

Linux 5.6 IO Scheduler BenchmarksOpenBenchmarking.orgPhoronix Test SuiteAMD Ryzen Threadripper 3970X 32-Core @ 3.70GHz (32 Cores / 64 Threads)ASUS ROG ZENITH II EXTREME (0702 BIOS)AMD Starship/Matisse64GB1000GB Force MP600Gigabyte AMD Radeon RX 5500/5500M / Pro 5500M 8GB (1890/875MHz)AMD Navi 10 HDMI AudioASUS MG28UAquantia AQC107 NBase-T/IEEE + Intel I211 + Intel Wi-Fi 6 AX200Ubuntu 20.045.6.0-050600rc7-generic (x86_64) 20200322GNOME Shell 3.35.91X Server 1.20.7modesetting 1.20.74.6 Mesa 20.0.0 (LLVM 9.0.1)GCC 9.3.0ext41280x8003840x2160ProcessorMotherboardChipsetMemoryDiskGraphicsAudioMonitorNetworkOSKernelDesktopDisplay ServerDisplay DriverOpenGLCompilerFile-SystemScreen ResolutionsLinux 5.6 IO Scheduler Benchmarks PerformanceSystem Logs- --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 - None: NONE / errors=remount-ro,relatime,rw- MQ-Deadline: MQ-DEADLINE / errors=remount-ro,relatime,rw- Kyber: KYBER / errors=remount-ro,relatime,rw- BFQ low-latency: BFQ / errors=remount-ro,relatime,rw- BFQ: BFQ / errors=remount-ro,relatime,rw- Scaling Governor: acpi-cpufreq ondemand - CPU Microcode: 0x8301025- OpenJDK Runtime Environment (build 11.0.6+10-post-Ubuntu-2ubuntu2)- itlb_multihit: Not affected + 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 Full AMD retpoline IBPB: conditional STIBP: conditional RSB filling + tsx_async_abort: Not affected

NoneMQ-DeadlineKyberBFQ low-latencyBFQLogarithmic Result OverviewPhoronix Test SuiteMariaDBApache CassandraPostgreSQL pgbenchFS-MarkSQLiteFlexible IO TesterFacebook RocksDBApache HBaseIORKeyDBTimed GDB GNU Debugger CompilationLevelDBTimed GCC Compilation

Linux 5.6 IO Scheduler Benchmarksmysqlslap: 64cassandra: Mixed 1:1build-gcc: Time To Compilesqlite: 8hbase: Async Rand Write - 32hbase: Async Rand Write - 32pgbench: Buffer Test - Normal Load - Read Writecassandra: Writesrocksdb: Rand Fill Synchbase: Seq Read - 32hbase: Seq Read - 32fio: Seq Write - IO_uring - Yes - No - 2MB - Default Test Directoryfio: Seq Write - IO_uring - Yes - No - 2MB - Default Test Directoryleveldb: Rand Deleteleveldb: Seq Fillleveldb: Seq Fillfs-mark: 5000 Files, 1MB Size, 4 Threadsfio: Rand Write - IO_uring - Yes - No - 4KB - Default Test Directoryfio: Rand Write - IO_uring - Yes - No - 4KB - Default Test Directorybuild-gdb: Time To Compilerocksdb: Rand Readkeydb: rocksdb: Read While Writingleveldb: Seek Randhbase: Increment - 32hbase: Increment - 32rocksdb: Rand Fillrocksdb: Seq Fillfio: Seq Write - IO_uring - Yes - No - 4KB - Default Test Directoryfio: Seq Write - IO_uring - Yes - No - 4KB - Default Test Directoryfio: Rand Read - IO_uring - Yes - No - 4KB - Default Test Directoryfio: Rand Read - IO_uring - Yes - No - 4KB - Default Test Directoryleveldb: Rand Readhbase: Async Rand Read - 32hbase: Async Rand Read - 32leveldb: Hot Readfio: Seq Read - IO_uring - Yes - No - 4KB - Default Test Directoryfio: Seq Read - IO_uring - Yes - No - 4KB - Default Test Directorysqlite: 1fio: Seq Read - IO_uring - Yes - No - 2MB - Default Test Directoryfio: Seq Read - IO_uring - Yes - No - 2MB - Default Test Directoryhbase: Rand Read - 32hbase: Rand Read - 32leveldb: Rand Fillleveldb: Rand Fillleveldb: Overwriteleveldb: Overwritefs-mark: 4000 Files, 32 Sub Dirs, 1MB Sizeleveldb: Fill Syncleveldb: Fill Syncior: Read Testior: Write TestNoneMQ-DeadlineKyberBFQ low-latencyBFQ11343753641.99044.7025226137234866.082212291627340821721853459861979256.304257.33127.51238.225066797980.411181322024519045.76616453566.582287110680120514512355583030001184353667138141.26718816737140.962379333148314.42319093825131239965265.43926.7264.66126.7592.31954.3763.6995.19797.73327226641.40865.0655016370612530.752488271238312091871701608471701257.507257.85427.4773.524946797580.246179996909522808.47618428866.890286110917118556312378222986671166286333111941.29519016600041.018426000166514.38519803967131239244265.97526.6265.35426.7591.41966.9123.6977.23794.35217165642.181142.9425356019712539.317088279252337151911662038841776257.428256.08427.6593.522614288380.361183324454527545.5613153066.609284111460119302012160322960001155285167111441.05417917574640.894493333192715.07620074021130240821265.71926.6265.71526.6545.81967.1613.6977.87781.43318141642.189151.3636954646112441.077629273739297611791774606861378257.174256.94227.5647.224075094080.893180517371525133.31611619066.22928811019611887471220974260000101520160078740.81119416258141.054351333137314.71818933792130241114264.86626.7265.54726.6590.61998.4353.5978.41799.22207137642.042107.8145086276112284.347087282883232751841733547031414258.148257.01427.5364.524575096080.797184232888522119.62629421066.34028411173011785581225970263000102720450080041.81919316360241.064492000192114.59820014009132238769265.21226.7264.24526.8528.42019.6663.5986.32770.48OpenBenchmarking.org

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 10.3.8Clients: 64NoneMQ-DeadlineBFQ low-latencyKyberBFQ2004006008001000SE +/- 1.14, N = 3SE +/- 101.84, N = 9SE +/- 100.08, N = 9SE +/- 2.81, N = 3SE +/- 3.55, N = 311343273182172071. (CXX) g++ options: -pie -fPIC -fstack-protector -fno-rtti -O2 -lpthread -lsnappy -laio -lz -lm -lcrypt -lssl -lcrypto -ldl
OpenBenchmarking.orgQueries Per Second, More Is BetterMariaDB 10.3.8Clients: 64NoneMQ-DeadlineBFQ low-latencyKyberBFQ2004006008001000Min: 1132.5 / Avg: 1134.22 / Max: 1136.36Min: 201.29 / Avg: 327.38 / Max: 1132.5Min: 199.64 / Avg: 317.92 / Max: 1117.32Min: 211.46 / Avg: 216.62 / Max: 221.14Min: 200.8 / Avg: 206.73 / Max: 213.081. (CXX) g++ options: -pie -fPIC -fstack-protector -fno-rtti -O2 -lpthread -lsnappy -laio -lz -lm -lcrypt -lssl -lcrypto -ldl

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: Mixed 1:1NoneMQ-DeadlineKyberBFQ low-latencyBFQ8001600240032004000SE +/- 217.63, N = 9SE +/- 5.46, N = 9SE +/- 12.43, N = 7SE +/- 26.61, N = 6SE +/- 8.58, N = 63753226165141137
OpenBenchmarking.orgOp/s, More Is BetterApache Cassandra 3.11.4Test: Mixed 1:1NoneMQ-DeadlineKyberBFQ low-latencyBFQ7001400210028003500Min: 3055 / Avg: 3753.11 / Max: 5021Min: 207 / Avg: 226.11 / Max: 263Min: 118 / Avg: 164.57 / Max: 216Min: 12 / Avg: 140.5 / Max: 188Min: 109 / Avg: 136.83 / Max: 163

Timed GCC Compilation

This test times how long it takes to build the GNU Compiler Collection (GCC). Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterTimed GCC Compilation 9.3.0Time To CompileMQ-DeadlineNoneBFQKyberBFQ low-latency140280420560700SE +/- 0.25, N = 3SE +/- 0.17, N = 3SE +/- 0.27, N = 3SE +/- 0.03, N = 3SE +/- 0.16, N = 3641.41641.99642.04642.18642.19
OpenBenchmarking.orgSeconds, Fewer Is BetterTimed GCC Compilation 9.3.0Time To CompileMQ-DeadlineNoneBFQKyberBFQ low-latency110220330440550Min: 640.97 / Avg: 641.41 / Max: 641.84Min: 641.8 / Avg: 641.99 / Max: 642.33Min: 641.56 / Avg: 642.04 / Max: 642.49Min: 642.13 / Avg: 642.18 / Max: 642.25Min: 641.96 / Avg: 642.19 / Max: 642.51

SQLite

This is a simple benchmark of SQLite. At present this test profile just measures the time to perform a pre-defined number of insertions on an indexed database. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterSQLite 3.30.1Threads / Copies: 8NoneMQ-DeadlineBFQKyberBFQ low-latency306090120150SE +/- 0.17, N = 3SE +/- 4.78, N = 15SE +/- 5.70, N = 15SE +/- 7.59, N = 12SE +/- 7.47, N = 1244.7065.07107.81142.94151.361. (CC) gcc options: -O2 -lz -lm -ldl -lpthread
OpenBenchmarking.orgSeconds, Fewer Is BetterSQLite 3.30.1Threads / Copies: 8NoneMQ-DeadlineBFQKyberBFQ low-latency306090120150Min: 44.44 / Avg: 44.7 / Max: 45.02Min: 46.16 / Avg: 65.07 / Max: 101.56Min: 53.98 / Avg: 107.81 / Max: 135.93Min: 98.66 / Avg: 142.94 / Max: 176.98Min: 119.9 / Avg: 151.36 / Max: 217.631. (CC) gcc options: -O2 -lz -lm -ldl -lpthread

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 2.2.3Test: Async Random Write - Clients: 32MQ-DeadlineBFQNoneKyberBFQ low-latency150300450600750SE +/- 6.18, N = 12SE +/- 3.48, N = 3SE +/- 13.83, N = 12SE +/- 16.45, N = 15SE +/- 24.46, N = 12501508522535695
OpenBenchmarking.orgMicroseconds - Average Latency, Fewer Is BetterApache HBase 2.2.3Test: Async Random Write - Clients: 32MQ-DeadlineBFQNoneKyberBFQ low-latency120240360480600Min: 491 / Avg: 500.75 / Max: 568Min: 502 / Avg: 507.67 / Max: 514Min: 499 / Avg: 522.42 / Max: 674Min: 494 / Avg: 535.27 / Max: 692Min: 599 / Avg: 695.25 / Max: 836

OpenBenchmarking.orgRows Per Second, More Is BetterApache HBase 2.2.3Test: Async Random Write - Clients: 32MQ-DeadlineBFQNoneKyberBFQ low-latency14K28K42K56K70KSE +/- 701.22, N = 12SE +/- 432.26, N = 3SE +/- 1286.78, N = 12SE +/- 1589.78, N = 15SE +/- 1485.96, N = 126370662761613726019746461
OpenBenchmarking.orgRows Per Second, More Is BetterApache HBase 2.2.3Test: Async Random Write - Clients: 32MQ-DeadlineBFQNoneKyberBFQ low-latency11K22K33K44K55KMin: 56085 / Avg: 63706.42 / Max: 64795Min: 61961 / Avg: 62760.67 / Max: 63445Min: 47314 / Avg: 61371.92 / Max: 63897Min: 46132 / Avg: 60196.6 / Max: 64492Min: 38207 / Avg: 46461.25 / Max: 53245

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 WriteNoneKyberMQ-DeadlineBFQ low-latencyBFQ7K14K21K28K35KSE +/- 238.48, N = 3SE +/- 139.65, N = 3SE +/- 170.74, N = 15SE +/- 212.09, N = 15SE +/- 162.47, N = 1534866.0812539.3212530.7512441.0812284.351. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -lcrypt -ldl -lm
OpenBenchmarking.orgTPS, More Is BetterPostgreSQL pgbench 12.0Scaling: Buffer Test - Test: Normal Load - Mode: Read WriteNoneKyberMQ-DeadlineBFQ low-latencyBFQ6K12K18K24K30KMin: 34472.99 / Avg: 34866.08 / Max: 35296.58Min: 12276.07 / Avg: 12539.32 / Max: 12751.77Min: 11359.55 / Avg: 12530.75 / Max: 13804.94Min: 10767.64 / Avg: 12441.08 / Max: 13838.27Min: 10903.68 / Avg: 12284.35 / Max: 13230.541. (CC) gcc options: -fno-strict-aliasing -fwrapv -O2 -lpgcommon -lpgport -lpq -lpthread -lrt -lcrypt -ldl -lm

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: WritesNoneBFQKyberBFQ low-latencyMQ-Deadline60K120K180K240K300KSE +/- 4255.75, N = 3SE +/- 3156.44, N = 15SE +/- 2588.05, N = 15SE +/- 1752.30, N = 3SE +/- 3838.79, N = 3291627282883279252273739271238
OpenBenchmarking.orgOp/s, More Is BetterApache Cassandra 3.11.4Test: WritesNoneBFQKyberBFQ low-latencyMQ-Deadline50K100K150K200K250KMin: 283117 / Avg: 291627.33 / Max: 296005Min: 263491 / Avg: 282882.67 / Max: 299242Min: 265176 / Avg: 279252.33 / Max: 292634Min: 271661 / Avg: 273739 / Max: 277222Min: 264111 / Avg: 271238 / Max: 277274

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 Fill SyncNoneKyberMQ-DeadlineBFQ low-latencyBFQ7K14K21K28K35KSE +/- 159.25, N = 3SE +/- 138.08, N = 3SE +/- 842.51, N = 12SE +/- 810.82, N = 15SE +/- 1587.48, N = 1234082337153120929761232751. (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 SyncNoneKyberMQ-DeadlineBFQ low-latencyBFQ6K12K18K24K30KMin: 33772 / Avg: 34081.67 / Max: 34301Min: 33439 / Avg: 33715 / Max: 33861Min: 25233 / Avg: 31209.42 / Max: 33819Min: 23299 / Avg: 29761.2 / Max: 32907Min: 6684 / Avg: 23274.58 / Max: 270131. (CXX) g++ options: -O3 -march=native -std=c++11 -fno-builtin-memcmp -fno-rtti -rdynamic -lpthread

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 2.2.3Test: Sequential Read - Clients: 32NoneBFQ low-latencyBFQMQ-DeadlineKyber4080120160200SE +/- 2.19, N = 15SE +/- 2.20, N = 15SE +/- 1.54, N = 15SE +/- 2.37, N = 15SE +/- 3.21, N = 3172179184187191
OpenBenchmarking.orgMicroseconds - Average Latency, Fewer Is BetterApache HBase 2.2.3Test: Sequential Read - Clients: 32NoneBFQ low-latencyBFQMQ-DeadlineKyber4080120160200Min: 167 / Avg: 171.73 / Max: 193Min: 171 / Avg: 179.4 / Max: 198Min: 181 / Avg: 184.4 / Max: 204Min: 175 / Avg: 187.47 / Max: 207Min: 185 / Avg: 191 / Max: 196

OpenBenchmarking.orgRows Per Second, More Is BetterApache HBase 2.2.3Test: Sequential Read - Clients: 32NoneBFQ low-latencyBFQMQ-DeadlineKyber40K80K120K160K200KSE +/- 2136.52, N = 15SE +/- 2066.20, N = 15SE +/- 1398.50, N = 15SE +/- 2100.10, N = 15SE +/- 2793.36, N = 3185345177460173354170160166203
OpenBenchmarking.orgRows Per Second, More Is BetterApache HBase 2.2.3Test: Sequential Read - Clients: 32NoneBFQ low-latencyBFQMQ-DeadlineKyber30K60K90K120K150KMin: 164556 / Avg: 185344.93 / Max: 190251Min: 160563 / Avg: 177459.8 / Max: 185791Min: 155558 / Avg: 173353.73 / Max: 176725Min: 153602 / Avg: 170159.6 / Max: 182041Min: 161775 / Avg: 166202.67 / Max: 171367

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: Sequential Write - Engine: IO_uring - Buffered: Yes - Direct: No - Block Size: 2MB - Disk Target: Default Test DirectoryNoneKyberMQ-DeadlineBFQBFQ low-latency2004006008001000SE +/- 10.37, N = 8SE +/- 36.75, N = 12SE +/- 27.25, N = 15SE +/- 49.26, N = 15SE +/- 68.36, N = 129868848477036861. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -march=native -ll -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 DirectoryNoneKyberMQ-DeadlineBFQBFQ low-latency2004006008001000Min: 916 / Avg: 986 / Max: 1007Min: 538 / Avg: 884.25 / Max: 974Min: 681 / Avg: 846.93 / Max: 959Min: 264 / Avg: 703.4 / Max: 817Min: 135 / Avg: 685.5 / Max: 8551. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -march=native -ll -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 DirectoryNoneKyberMQ-DeadlineBFQBFQ low-latency400800120016002000SE +/- 20.59, N = 8SE +/- 73.55, N = 12SE +/- 54.46, N = 15SE +/- 98.56, N = 15SE +/- 136.82, N = 12197917761701141413781. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -march=native -ll -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 DirectoryNoneKyberMQ-DeadlineBFQBFQ low-latency30060090012001500Min: 1840 / Avg: 1979 / Max: 2020Min: 1082 / Avg: 1775.58 / Max: 1955Min: 1370 / Avg: 1701.33 / Max: 1925Min: 535 / Avg: 1413.93 / Max: 1641Min: 276 / Avg: 1378.25 / Max: 17181. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -march=native -ll -lrt -laio -lz -lpthread -lm -ldl

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: Random DeleteNoneBFQ low-latencyKyberMQ-DeadlineBFQ60120180240300SE +/- 0.66, N = 3SE +/- 0.13, N = 3SE +/- 0.37, N = 3SE +/- 0.58, N = 3SE +/- 0.64, N = 3256.30257.17257.43257.51258.151. (CXX) g++ options: -O3 -lsnappy -lpthread
OpenBenchmarking.orgMicroseconds Per Op, Fewer Is BetterLevelDB 1.22Benchmark: Random DeleteNoneBFQ low-latencyKyberMQ-DeadlineBFQ50100150200250Min: 255.37 / Avg: 256.3 / Max: 257.58Min: 256.97 / Avg: 257.17 / Max: 257.42Min: 256.74 / Avg: 257.43 / Max: 258.02Min: 256.77 / Avg: 257.51 / Max: 258.65Min: 257.07 / Avg: 258.15 / Max: 259.291. (CXX) g++ options: -O3 -lsnappy -lpthread

OpenBenchmarking.orgMicroseconds Per Op, Fewer Is BetterLevelDB 1.22Benchmark: Sequential FillKyberBFQ low-latencyBFQNoneMQ-Deadline60120180240300SE +/- 0.53, N = 3SE +/- 0.62, N = 3SE +/- 0.62, N = 3SE +/- 0.55, N = 3SE +/- 0.41, N = 3256.08256.94257.01257.33257.851. (CXX) g++ options: -O3 -lsnappy -lpthread
OpenBenchmarking.orgMicroseconds Per Op, Fewer Is BetterLevelDB 1.22Benchmark: Sequential FillKyberBFQ low-latencyBFQNoneMQ-Deadline50100150200250Min: 255.32 / Avg: 256.08 / Max: 257.11Min: 256.01 / Avg: 256.94 / Max: 258.12Min: 255.82 / Avg: 257.01 / Max: 257.91Min: 256.3 / Avg: 257.33 / Max: 258.18Min: 257.06 / Avg: 257.85 / Max: 258.461. (CXX) g++ options: -O3 -lsnappy -lpthread

OpenBenchmarking.orgMB/s, More Is BetterLevelDB 1.22Benchmark: Sequential FillKyberBFQBFQ low-latencyNoneMQ-Deadline612182430SE +/- 0.07, N = 3SE +/- 0.09, N = 3SE +/- 0.07, N = 3SE +/- 0.06, N = 3SE +/- 0.03, N = 327.627.527.527.527.41. (CXX) g++ options: -O3 -lsnappy -lpthread
OpenBenchmarking.orgMB/s, More Is BetterLevelDB 1.22Benchmark: Sequential FillKyberBFQBFQ low-latencyNoneMQ-Deadline612182430Min: 27.5 / Avg: 27.63 / Max: 27.7Min: 27.4 / Avg: 27.53 / Max: 27.7Min: 27.4 / Avg: 27.53 / Max: 27.6Min: 27.4 / Avg: 27.5 / Max: 27.6Min: 27.4 / Avg: 27.43 / Max: 27.51. (CXX) g++ options: -O3 -lsnappy -lpthread

FS-Mark

FS_Mark is designed to test a system's file-system performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFiles/s, More Is BetterFS-Mark 3.3Test: 5000 Files, 1MB Size, 4 ThreadsNoneMQ-DeadlineBFQ low-latencyKyberBFQ30060090012001500SE +/- 9.23, N = 3SE +/- 121.16, N = 12SE +/- 103.97, N = 12SE +/- 87.62, N = 12SE +/- 2.86, N = 31238.2773.5647.2593.5364.51. (CC) gcc options: -static
OpenBenchmarking.orgFiles/s, More Is BetterFS-Mark 3.3Test: 5000 Files, 1MB Size, 4 ThreadsNoneMQ-DeadlineBFQ low-latencyKyberBFQ2004006008001000Min: 1219.9 / Avg: 1238.17 / Max: 1249.6Min: 369.2 / Avg: 773.54 / Max: 1251.2Min: 368.5 / Avg: 647.21 / Max: 1238.7Min: 339.4 / Avg: 593.48 / Max: 1057.1Min: 359.5 / Avg: 364.53 / Max: 369.41. (CC) gcc options: -static

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: 4KB - Disk Target: Default Test DirectoryNoneMQ-DeadlineBFQBFQ low-latencyKyber50K100K150K200K250KSE +/- 2603.42, N = 3SE +/- 2212.23, N = 15SE +/- 4178.09, N = 12SE +/- 3250.00, N = 4SE +/- 14652.50, N = 122506672494672457502407502261421. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -march=native -ll -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 DirectoryNoneMQ-DeadlineBFQBFQ low-latencyKyber40K80K120K160K200KMin: 246000 / Avg: 250666.67 / Max: 255000Min: 226000 / Avg: 249466.67 / Max: 259000Min: 205000 / Avg: 245750 / Max: 260000Min: 232000 / Avg: 240750 / Max: 247000Min: 75700 / Avg: 226141.67 / Max: 2530001. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -march=native -ll -lrt -laio -lz -lpthread -lm -ldl

OpenBenchmarking.orgMB/s, More Is BetterFlexible IO Tester 3.18Type: Random Write - Engine: IO_uring - Buffered: Yes - Direct: No - Block Size: 4KB - Disk Target: Default Test DirectoryNoneMQ-DeadlineBFQBFQ low-latencyKyber2004006008001000SE +/- 10.41, N = 3SE +/- 8.60, N = 15SE +/- 16.31, N = 12SE +/- 12.91, N = 4SE +/- 57.22, N = 129799759609408831. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -march=native -ll -lrt -laio -lz -lpthread -lm -ldl
OpenBenchmarking.orgMB/s, More Is BetterFlexible IO Tester 3.18Type: Random Write - Engine: IO_uring - Buffered: Yes - Direct: No - Block Size: 4KB - Disk Target: Default Test DirectoryNoneMQ-DeadlineBFQBFQ low-latencyKyber2004006008001000Min: 960 / Avg: 978.67 / Max: 996Min: 884 / Avg: 974.93 / Max: 1013Min: 801 / Avg: 959.67 / Max: 1017Min: 905 / Avg: 940.25 / Max: 964Min: 296 / Avg: 883.42 / Max: 9881. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -march=native -ll -lrt -laio -lz -lpthread -lm -ldl

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 CompileMQ-DeadlineKyberNoneBFQBFQ low-latency20406080100SE +/- 0.08, N = 3SE +/- 0.06, N = 3SE +/- 0.11, N = 3SE +/- 0.05, N = 3SE +/- 0.05, N = 380.2580.3680.4180.8080.89
OpenBenchmarking.orgSeconds, Fewer Is BetterTimed GDB GNU Debugger Compilation 9.1Time To CompileMQ-DeadlineKyberNoneBFQBFQ low-latency1530456075Min: 80.14 / Avg: 80.25 / Max: 80.41Min: 80.28 / Avg: 80.36 / Max: 80.47Min: 80.29 / Avg: 80.41 / Max: 80.63Min: 80.72 / Avg: 80.8 / Max: 80.9Min: 80.8 / Avg: 80.89 / Max: 80.96

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 ReadBFQKyberNoneBFQ low-latencyMQ-Deadline40M80M120M160M200MSE +/- 2076338.73, N = 6SE +/- 3150233.61, N = 3SE +/- 2048541.15, N = 3SE +/- 2267118.09, N = 5SE +/- 1286835.14, N = 31842328881833244541813220241805173711799969091. (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 ReadBFQKyberNoneBFQ low-latencyMQ-Deadline30M60M90M120M150MMin: 174502992 / Avg: 184232887.83 / Max: 188344292Min: 177512979 / Avg: 183324454 / Max: 188337793Min: 177243498 / Avg: 181322024.33 / Max: 183698600Min: 176243636 / Avg: 180517371 / Max: 188230830Min: 177667589 / Avg: 179996909 / Max: 1821094801. (CXX) g++ options: -O3 -march=native -std=c++11 -fno-builtin-memcmp -fno-rtti -rdynamic -lpthread

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 5.3.1KyberBFQ low-latencyMQ-DeadlineBFQNone110K220K330K440K550KSE +/- 2621.85, N = 3SE +/- 2938.61, N = 3SE +/- 5087.07, N = 3SE +/- 3207.20, N = 3SE +/- 3949.59, N = 3527545.50525133.31522808.47522119.62519045.761. (CXX) g++ options: -O2 -levent -lpthread -lz -lpcre
OpenBenchmarking.orgOps/sec, More Is BetterKeyDB 5.3.1KyberBFQ low-latencyMQ-DeadlineBFQNone90K180K270K360K450KMin: 522319.5 / Avg: 527545.53 / Max: 530531Min: 519519.67 / Avg: 525133.31 / Max: 529447.29Min: 512672.05 / Avg: 522808.47 / Max: 528634.69Min: 517829.76 / Avg: 522119.62 / Max: 528394.46Min: 512975.13 / Avg: 519045.76 / Max: 526458.111. (CXX) g++ options: -O2 -levent -lpthread -lz -lpcre

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: Read While WritingBFQMQ-DeadlineNoneKyberBFQ low-latency1.3M2.6M3.9M5.2M6.5MSE +/- 70731.04, N = 3SE +/- 80811.91, N = 5SE +/- 29916.58, N = 3SE +/- 21536.07, N = 3SE +/- 22104.21, N = 3629421061842886164535613153061161901. (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 WritingBFQMQ-DeadlineNoneKyberBFQ low-latency1.1M2.2M3.3M4.4M5.5MMin: 6213994 / Avg: 6294209.67 / Max: 6435227Min: 6040407 / Avg: 6184287.6 / Max: 6496382Min: 6105253 / Avg: 6164535 / Max: 6201193Min: 6099350 / Avg: 6131530 / Max: 6172414Min: 6072042 / Avg: 6116189.67 / Max: 61402701. (CXX) g++ options: -O3 -march=native -std=c++11 -fno-builtin-memcmp -fno-rtti -rdynamic -lpthread

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: Seek RandomBFQ low-latencyBFQNoneKyberMQ-Deadline1530456075SE +/- 0.24, N = 3SE +/- 0.07, N = 3SE +/- 0.15, N = 3SE +/- 0.11, N = 3SE +/- 0.04, N = 366.2366.3466.5866.6166.891. (CXX) g++ options: -O3 -lsnappy -lpthread
OpenBenchmarking.orgMicroseconds Per Op, Fewer Is BetterLevelDB 1.22Benchmark: Seek RandomBFQ low-latencyBFQNoneKyberMQ-Deadline1326395265Min: 65.8 / Avg: 66.23 / Max: 66.63Min: 66.2 / Avg: 66.34 / Max: 66.42Min: 66.3 / Avg: 66.58 / Max: 66.81Min: 66.4 / Avg: 66.61 / Max: 66.76Min: 66.81 / Avg: 66.89 / Max: 66.941. (CXX) g++ options: -O3 -lsnappy -lpthread

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 2.2.3Test: Increment - Clients: 32KyberBFQMQ-DeadlineNoneBFQ low-latency60120180240300SE +/- 1.86, N = 3SE +/- 2.89, N = 3SE +/- 2.19, N = 3SE +/- 1.45, N = 3284284286287288
OpenBenchmarking.orgMicroseconds - Average Latency, Fewer Is BetterApache HBase 2.2.3Test: Increment - Clients: 32KyberBFQMQ-DeadlineNoneBFQ low-latency50100150200250Min: 282 / Avg: 284.33 / Max: 288Min: 279 / Avg: 284 / Max: 289Min: 283 / Avg: 285.67 / Max: 290Min: 284 / Avg: 286.67 / Max: 289

OpenBenchmarking.orgRows Per Second, More Is BetterApache HBase 2.2.3Test: Increment - Clients: 32BFQKyberMQ-DeadlineNoneBFQ low-latency20K40K60K80K100KSE +/- 1065.86, N = 3SE +/- 732.48, N = 3SE +/- 672.36, N = 3SE +/- 367.77, N = 3SE +/- 503.81, N = 3111730111460110917110680110196
OpenBenchmarking.orgRows Per Second, More Is BetterApache HBase 2.2.3Test: Increment - Clients: 32BFQKyberMQ-DeadlineNoneBFQ low-latency20K40K60K80K100KMin: 109830 / Avg: 111730.33 / Max: 113517Min: 109997 / Avg: 111459.67 / Max: 112262Min: 109572 / Avg: 110916.67 / Max: 111599Min: 110044 / Avg: 110680 / Max: 111318Min: 109403 / Avg: 110196.33 / Max: 111131

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 FillNoneKyberBFQ low-latencyMQ-DeadlineBFQ300K600K900K1200K1500KSE +/- 3276.95, N = 3SE +/- 2916.55, N = 3SE +/- 2824.61, N = 3SE +/- 11269.49, N = 3SE +/- 5909.18, N = 3120514511930201188747118556311785581. (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 FillNoneKyberBFQ low-latencyMQ-DeadlineBFQ200K400K600K800K1000KMin: 1198745 / Avg: 1205145.33 / Max: 1209567Min: 1187396 / Avg: 1193019.67 / Max: 1197173Min: 1185102 / Avg: 1188746.67 / Max: 1194307Min: 1166371 / Avg: 1185563.33 / Max: 1205394Min: 1171280 / Avg: 1178558 / Max: 11902611. (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 FillMQ-DeadlineNoneBFQBFQ low-latencyKyber300K600K900K1200K1500KSE +/- 3005.09, N = 3SE +/- 9824.57, N = 3SE +/- 11522.86, N = 3SE +/- 3978.59, N = 3SE +/- 9197.27, N = 3123782212355581225970122097412160321. (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 FillMQ-DeadlineNoneBFQBFQ low-latencyKyber200K400K600K800K1000KMin: 1234749 / Avg: 1237822.33 / Max: 1243832Min: 1216083 / Avg: 1235558 / Max: 1247556Min: 1211678 / Avg: 1225970.33 / Max: 1248773Min: 1213249 / Avg: 1220974 / Max: 1226489Min: 1204568 / Avg: 1216032 / Max: 12342221. (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: Sequential Write - Engine: IO_uring - Buffered: Yes - Direct: No - Block Size: 4KB - Disk Target: Default Test DirectoryNoneMQ-DeadlineKyberBFQBFQ low-latency60K120K180K240K300KSE +/- 2679.89, N = 11SE +/- 3179.80, N = 3SE +/- 2309.40, N = 3SE +/- 3785.94, N = 3SE +/- 3511.88, N = 33030002986672960002630002600001. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -march=native -ll -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: 4KB - Disk Target: Default Test DirectoryNoneMQ-DeadlineKyberBFQBFQ low-latency50K100K150K200K250KMin: 289000 / Avg: 303000 / Max: 315000Min: 295000 / Avg: 298666.67 / Max: 305000Min: 292000 / Avg: 296000 / Max: 300000Min: 256000 / Avg: 263000 / Max: 269000Min: 253000 / Avg: 260000 / Max: 2640001. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -march=native -ll -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: 4KB - Disk Target: Default Test DirectoryNoneMQ-DeadlineKyberBFQBFQ low-latency30060090012001500SE +/- 10.45, N = 11SE +/- 11.84, N = 3SE +/- 8.11, N = 3SE +/- 15.25, N = 3SE +/- 13.20, N = 3118411661155102710151. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -march=native -ll -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: 4KB - Disk Target: Default Test DirectoryNoneMQ-DeadlineKyberBFQBFQ low-latency2004006008001000Min: 1130 / Avg: 1183.55 / Max: 1232Min: 1154 / Avg: 1166.33 / Max: 1190Min: 1142 / Avg: 1155.33 / Max: 1170Min: 998 / Avg: 1026.67 / Max: 1050Min: 989 / Avg: 1015.33 / Max: 10301. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -march=native -ll -lrt -laio -lz -lpthread -lm -ldl

OpenBenchmarking.orgIOPS, More Is BetterFlexible IO Tester 3.18Type: Random Read - Engine: IO_uring - Buffered: Yes - Direct: No - Block Size: 4KB - Disk Target: Default Test DirectoryNoneMQ-DeadlineKyberBFQBFQ low-latency80K160K240K320K400KSE +/- 1452.97, N = 3SE +/- 1763.83, N = 3SE +/- 3439.15, N = 6SE +/- 2843.12, N = 4SE +/- 2481.93, N = 53536672863332851672045002016001. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -march=native -ll -lrt -laio -lz -lpthread -lm -ldl
OpenBenchmarking.orgIOPS, More Is BetterFlexible IO Tester 3.18Type: Random Read - Engine: IO_uring - Buffered: Yes - Direct: No - Block Size: 4KB - Disk Target: Default Test DirectoryNoneMQ-DeadlineKyberBFQBFQ low-latency60K120K180K240K300KMin: 351000 / Avg: 353666.67 / Max: 356000Min: 283000 / Avg: 286333.33 / Max: 289000Min: 268000 / Avg: 285166.67 / Max: 289000Min: 196000 / Avg: 204500 / Max: 208000Min: 192000 / Avg: 201600 / Max: 2060001. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -march=native -ll -lrt -laio -lz -lpthread -lm -ldl

OpenBenchmarking.orgMB/s, More Is BetterFlexible IO Tester 3.18Type: Random Read - Engine: IO_uring - Buffered: Yes - Direct: No - Block Size: 4KB - Disk Target: Default Test DirectoryNoneMQ-DeadlineKyberBFQBFQ low-latency30060090012001500SE +/- 5.78, N = 3SE +/- 6.17, N = 3SE +/- 13.18, N = 6SE +/- 11.02, N = 4SE +/- 9.65, N = 51381111911148007871. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -march=native -ll -lrt -laio -lz -lpthread -lm -ldl
OpenBenchmarking.orgMB/s, More Is BetterFlexible IO Tester 3.18Type: Random Read - Engine: IO_uring - Buffered: Yes - Direct: No - Block Size: 4KB - Disk Target: Default Test DirectoryNoneMQ-DeadlineKyberBFQBFQ low-latency2004006008001000Min: 1371 / Avg: 1381.33 / Max: 1391Min: 1107 / Avg: 1118.67 / Max: 1128Min: 1048 / Avg: 1113.83 / Max: 1130Min: 767 / Avg: 800 / Max: 813Min: 750 / Avg: 787.2 / Max: 8041. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -march=native -ll -lrt -laio -lz -lpthread -lm -ldl

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: Random ReadBFQ low-latencyKyberNoneMQ-DeadlineBFQ1020304050SE +/- 0.15, N = 3SE +/- 0.09, N = 3SE +/- 0.15, N = 3SE +/- 0.18, N = 3SE +/- 0.44, N = 340.8141.0541.2741.3041.821. (CXX) g++ options: -O3 -lsnappy -lpthread
OpenBenchmarking.orgMicroseconds Per Op, Fewer Is BetterLevelDB 1.22Benchmark: Random ReadBFQ low-latencyKyberNoneMQ-DeadlineBFQ918273645Min: 40.58 / Avg: 40.81 / Max: 41.09Min: 40.89 / Avg: 41.05 / Max: 41.2Min: 41.03 / Avg: 41.27 / Max: 41.55Min: 41.02 / Avg: 41.3 / Max: 41.64Min: 41.26 / Avg: 41.82 / Max: 42.681. (CXX) g++ options: -O3 -lsnappy -lpthread

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 2.2.3Test: Async Random Read - Clients: 32KyberNoneMQ-DeadlineBFQBFQ low-latency4080120160200SE +/- 2.03, N = 3SE +/- 1.86, N = 3SE +/- 2.33, N = 3SE +/- 2.19, N = 3SE +/- 1.76, N = 3179188190193194
OpenBenchmarking.orgMicroseconds - Average Latency, Fewer Is BetterApache HBase 2.2.3Test: Async Random Read - Clients: 32KyberNoneMQ-DeadlineBFQBFQ low-latency4080120160200Min: 176 / Avg: 179.33 / Max: 183Min: 186 / Avg: 188.33 / Max: 192Min: 186 / Avg: 189.67 / Max: 194Min: 190 / Avg: 192.67 / Max: 197Min: 191 / Avg: 193.67 / Max: 197

OpenBenchmarking.orgRows Per Second, More Is BetterApache HBase 2.2.3Test: Async Random Read - Clients: 32KyberNoneMQ-DeadlineBFQBFQ low-latency40K80K120K160K200KSE +/- 2062.43, N = 3SE +/- 1741.28, N = 3SE +/- 1862.94, N = 3SE +/- 1905.12, N = 3SE +/- 1743.02, N = 3175746167371166000163602162581
OpenBenchmarking.orgRows Per Second, More Is BetterApache HBase 2.2.3Test: Async Random Read - Clients: 32KyberNoneMQ-DeadlineBFQBFQ low-latency30K60K90K120K150KMin: 172216 / Avg: 175745.67 / Max: 179359Min: 163953 / Avg: 167370.67 / Max: 169659Min: 162505 / Avg: 165999.67 / Max: 168866Min: 159836 / Avg: 163602.33 / Max: 165985Min: 159156 / Avg: 162581.33 / Max: 164855

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 ReadKyberNoneMQ-DeadlineBFQ low-latencyBFQ918273645SE +/- 0.14, N = 3SE +/- 0.21, N = 3SE +/- 0.16, N = 3SE +/- 0.18, N = 3SE +/- 0.26, N = 340.8940.9641.0241.0541.061. (CXX) g++ options: -O3 -lsnappy -lpthread
OpenBenchmarking.orgMicroseconds Per Op, Fewer Is BetterLevelDB 1.22Benchmark: Hot ReadKyberNoneMQ-DeadlineBFQ low-latencyBFQ918273645Min: 40.61 / Avg: 40.89 / Max: 41.07Min: 40.59 / Avg: 40.96 / Max: 41.32Min: 40.78 / Avg: 41.02 / Max: 41.32Min: 40.77 / Avg: 41.05 / Max: 41.38Min: 40.55 / Avg: 41.06 / Max: 41.321. (CXX) g++ options: -O3 -lsnappy -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: Sequential Read - Engine: IO_uring - Buffered: Yes - Direct: No - Block Size: 4KB - Disk Target: Default Test DirectoryKyberBFQMQ-DeadlineNoneBFQ low-latency110K220K330K440K550KSE +/- 1666.67, N = 3SE +/- 6670.83, N = 4SE +/- 3214.55, N = 3SE +/- 1452.97, N = 3SE +/- 4176.65, N = 34933334920004260003793333513331. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -march=native -ll -lrt -laio -lz -lpthread -lm -ldl
OpenBenchmarking.orgIOPS, More Is BetterFlexible IO Tester 3.18Type: Sequential Read - Engine: IO_uring - Buffered: Yes - Direct: No - Block Size: 4KB - Disk Target: Default Test DirectoryKyberBFQMQ-DeadlineNoneBFQ low-latency90K180K270K360K450KMin: 490000 / Avg: 493333.33 / Max: 495000Min: 472000 / Avg: 492000 / Max: 499000Min: 421000 / Avg: 426000 / Max: 432000Min: 377000 / Avg: 379333.33 / Max: 382000Min: 343000 / Avg: 351333.33 / Max: 3560001. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -march=native -ll -lrt -laio -lz -lpthread -lm -ldl

OpenBenchmarking.orgMB/s, More Is BetterFlexible IO Tester 3.18Type: Sequential Read - Engine: IO_uring - Buffered: Yes - Direct: No - Block Size: 4KB - Disk Target: Default Test DirectoryKyberBFQMQ-DeadlineNoneBFQ low-latency400800120016002000SE +/- 7.17, N = 3SE +/- 26.11, N = 4SE +/- 12.35, N = 3SE +/- 5.81, N = 3SE +/- 16.23, N = 3192719211665148313731. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -march=native -ll -lrt -laio -lz -lpthread -lm -ldl
OpenBenchmarking.orgMB/s, More Is BetterFlexible IO Tester 3.18Type: Sequential Read - Engine: IO_uring - Buffered: Yes - Direct: No - Block Size: 4KB - Disk Target: Default Test DirectoryKyberBFQMQ-DeadlineNoneBFQ low-latency30060090012001500Min: 1913 / Avg: 1927.33 / Max: 1935Min: 1843 / Avg: 1921.25 / Max: 1950Min: 1646 / Avg: 1664.67 / Max: 1688Min: 1474 / Avg: 1483.33 / Max: 1494Min: 1341 / Avg: 1373.33 / Max: 13921. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -march=native -ll -lrt -laio -lz -lpthread -lm -ldl

SQLite

This is a simple benchmark of SQLite. At present this test profile just measures the time to perform a pre-defined number of insertions on an indexed database. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgSeconds, Fewer Is BetterSQLite 3.30.1Threads / Copies: 1MQ-DeadlineNoneBFQBFQ low-latencyKyber48121620SE +/- 0.05, N = 3SE +/- 0.05, N = 3SE +/- 0.14, N = 3SE +/- 0.15, N = 8SE +/- 0.20, N = 1514.3914.4214.6014.7215.081. (CC) gcc options: -O2 -lz -lm -ldl -lpthread
OpenBenchmarking.orgSeconds, Fewer Is BetterSQLite 3.30.1Threads / Copies: 1MQ-DeadlineNoneBFQBFQ low-latencyKyber48121620Min: 14.29 / Avg: 14.39 / Max: 14.46Min: 14.35 / Avg: 14.42 / Max: 14.51Min: 14.44 / Avg: 14.6 / Max: 14.88Min: 14.41 / Avg: 14.72 / Max: 15.69Min: 14.55 / Avg: 15.08 / Max: 16.461. (CC) gcc options: -O2 -lz -lm -ldl -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: Sequential Read - Engine: IO_uring - Buffered: Yes - Direct: No - Block Size: 2MB - Disk Target: Default Test DirectoryKyberBFQMQ-DeadlineNoneBFQ low-latency400800120016002000SE +/- 2.40, N = 3SE +/- 0.67, N = 3SE +/- 1.15, N = 3SE +/- 1.67, N = 3SE +/- 1.33, N = 3200720011980190918931. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -march=native -ll -lrt -laio -lz -lpthread -lm -ldl
OpenBenchmarking.orgIOPS, More Is BetterFlexible IO Tester 3.18Type: Sequential Read - Engine: IO_uring - Buffered: Yes - Direct: No - Block Size: 2MB - Disk Target: Default Test DirectoryKyberBFQMQ-DeadlineNoneBFQ low-latency30060090012001500Min: 2004 / Avg: 2007.33 / Max: 2012Min: 2000 / Avg: 2000.67 / Max: 2002Min: 1978 / Avg: 1980 / Max: 1982Min: 1907 / Avg: 1908.67 / Max: 1912Min: 1890 / Avg: 1892.67 / Max: 18941. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -march=native -ll -lrt -laio -lz -lpthread -lm -ldl

OpenBenchmarking.orgMB/s, More Is BetterFlexible IO Tester 3.18Type: Sequential Read - Engine: IO_uring - Buffered: Yes - Direct: No - Block Size: 2MB - Disk Target: Default Test DirectoryKyberBFQMQ-DeadlineNoneBFQ low-latency9001800270036004500SE +/- 4.73, N = 3SE +/- 1.33, N = 3SE +/- 2.33, N = 3SE +/- 3.33, N = 3SE +/- 2.67, N = 3402140093967382537921. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -march=native -ll -lrt -laio -lz -lpthread -lm -ldl
OpenBenchmarking.orgMB/s, More Is BetterFlexible IO Tester 3.18Type: Sequential Read - Engine: IO_uring - Buffered: Yes - Direct: No - Block Size: 2MB - Disk Target: Default Test DirectoryKyberBFQMQ-DeadlineNoneBFQ low-latency7001400210028003500Min: 4014 / Avg: 4021 / Max: 4030Min: 4008 / Avg: 4009.33 / Max: 4012Min: 3963 / Avg: 3967.33 / Max: 3971Min: 3822 / Avg: 3825.33 / Max: 3832Min: 3787 / Avg: 3792.33 / Max: 37951. (CC) gcc options: -rdynamic -std=gnu99 -ffast-math -include -O3 -U_FORTIFY_SOURCE -march=native -ll -lrt -laio -lz -lpthread -lm -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 2.2.3Test: Random Read - Clients: 32KyberBFQ low-latencyNoneMQ-DeadlineBFQ306090120150SE +/- 0.67, N = 3SE +/- 0.67, N = 3130130131131132
OpenBenchmarking.orgMicroseconds - Average Latency, Fewer Is BetterApache HBase 2.2.3Test: Random Read - Clients: 32KyberBFQ low-latencyNoneMQ-DeadlineBFQ20406080100Min: 129 / Avg: 130.33 / Max: 131Min: 130 / Avg: 131.33 / Max: 132

OpenBenchmarking.orgRows Per Second, More Is BetterApache HBase 2.2.3Test: Random Read - Clients: 32BFQ low-latencyKyberNoneMQ-DeadlineBFQ50K100K150K200K250KSE +/- 1176.18, N = 3SE +/- 1147.34, N = 3SE +/- 1261.46, N = 3SE +/- 1696.17, N = 3SE +/- 1073.93, N = 3241114240821239965239244238769
OpenBenchmarking.orgRows Per Second, More Is BetterApache HBase 2.2.3Test: Random Read - Clients: 32BFQ low-latencyKyberNoneMQ-DeadlineBFQ40K80K120K160K200KMin: 239047 / Avg: 241114.33 / Max: 243120Min: 239563 / Avg: 240821 / Max: 243112Min: 238479 / Avg: 239965.33 / Max: 242474Min: 237000 / Avg: 239243.67 / Max: 242569Min: 237489 / Avg: 238769.33 / Max: 240903

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: Random FillBFQ low-latencyBFQNoneKyberMQ-Deadline60120180240300SE +/- 0.15, N = 3SE +/- 0.38, N = 3SE +/- 0.41, N = 3SE +/- 0.64, N = 3SE +/- 0.87, N = 3264.87265.21265.44265.72265.981. (CXX) g++ options: -O3 -lsnappy -lpthread
OpenBenchmarking.orgMicroseconds Per Op, Fewer Is BetterLevelDB 1.22Benchmark: Random FillBFQ low-latencyBFQNoneKyberMQ-Deadline50100150200250Min: 264.69 / Avg: 264.87 / Max: 265.16Min: 264.8 / Avg: 265.21 / Max: 265.97Min: 265 / Avg: 265.44 / Max: 266.27Min: 264.9 / Avg: 265.72 / Max: 266.97Min: 264.59 / Avg: 265.97 / Max: 267.61. (CXX) g++ options: -O3 -lsnappy -lpthread

OpenBenchmarking.orgMB/s, More Is BetterLevelDB 1.22Benchmark: Random FillBFQBFQ low-latencyNoneKyberMQ-Deadline612182430SE +/- 0.03, N = 3SE +/- 0.00, N = 3SE +/- 0.03, N = 3SE +/- 0.07, N = 3SE +/- 0.09, N = 326.726.726.726.626.61. (CXX) g++ options: -O3 -lsnappy -lpthread
OpenBenchmarking.orgMB/s, More Is BetterLevelDB 1.22Benchmark: Random FillBFQBFQ low-latencyNoneKyberMQ-Deadline612182430Min: 26.6 / Avg: 26.67 / Max: 26.7Min: 26.7 / Avg: 26.7 / Max: 26.7Min: 26.6 / Avg: 26.67 / Max: 26.7Min: 26.5 / Avg: 26.63 / Max: 26.7Min: 26.4 / Avg: 26.57 / Max: 26.71. (CXX) g++ options: -O3 -lsnappy -lpthread

OpenBenchmarking.orgMicroseconds Per Op, Fewer Is BetterLevelDB 1.22Benchmark: OverwriteBFQNoneMQ-DeadlineBFQ low-latencyKyber60120180240300SE +/- 0.58, N = 3SE +/- 0.82, N = 3SE +/- 0.58, N = 3SE +/- 0.56, N = 3SE +/- 1.16, N = 3264.25264.66265.35265.55265.721. (CXX) g++ options: -O3 -lsnappy -lpthread
OpenBenchmarking.orgMicroseconds Per Op, Fewer Is BetterLevelDB 1.22Benchmark: OverwriteBFQNoneMQ-DeadlineBFQ low-latencyKyber50100150200250Min: 263.12 / Avg: 264.25 / Max: 265.02Min: 263.09 / Avg: 264.66 / Max: 265.88Min: 264.2 / Avg: 265.35 / Max: 266.04Min: 264.61 / Avg: 265.55 / Max: 266.56Min: 263.91 / Avg: 265.71 / Max: 267.891. (CXX) g++ options: -O3 -lsnappy -lpthread

OpenBenchmarking.orgMB/s, More Is BetterLevelDB 1.22Benchmark: OverwriteBFQMQ-DeadlineNoneBFQ low-latencyKyber612182430SE +/- 0.07, N = 3SE +/- 0.07, N = 3SE +/- 0.09, N = 3SE +/- 0.06, N = 3SE +/- 0.12, N = 326.826.726.726.626.61. (CXX) g++ options: -O3 -lsnappy -lpthread
OpenBenchmarking.orgMB/s, More Is BetterLevelDB 1.22Benchmark: OverwriteBFQMQ-DeadlineNoneBFQ low-latencyKyber612182430Min: 26.7 / Avg: 26.77 / Max: 26.9Min: 26.6 / Avg: 26.67 / Max: 26.8Min: 26.6 / Avg: 26.73 / Max: 26.9Min: 26.5 / Avg: 26.6 / Max: 26.7Min: 26.4 / Avg: 26.63 / Max: 26.81. (CXX) g++ options: -O3 -lsnappy -lpthread

FS-Mark

FS_Mark is designed to test a system's file-system performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgFiles/s, More Is BetterFS-Mark 3.3Test: 4000 Files, 32 Sub Dirs, 1MB SizeNoneMQ-DeadlineBFQ low-latencyKyberBFQ130260390520650SE +/- 0.97, N = 3SE +/- 2.75, N = 3SE +/- 1.70, N = 3SE +/- 24.28, N = 13SE +/- 21.50, N = 12592.3591.4590.6545.8528.41. (CC) gcc options: -static
OpenBenchmarking.orgFiles/s, More Is BetterFS-Mark 3.3Test: 4000 Files, 32 Sub Dirs, 1MB SizeNoneMQ-DeadlineBFQ low-latencyKyberBFQ100200300400500Min: 590.4 / Avg: 592.33 / Max: 593.4Min: 586.2 / Avg: 591.43 / Max: 595.5Min: 588.2 / Avg: 590.63 / Max: 593.9Min: 302.5 / Avg: 545.78 / Max: 595.2Min: 331.4 / Avg: 528.38 / Max: 601.31. (CC) gcc options: -static

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: Fill SyncNoneMQ-DeadlineKyberBFQ low-latencyBFQ400800120016002000SE +/- 3.92, N = 3SE +/- 5.68, N = 3SE +/- 13.46, N = 3SE +/- 19.30, N = 3SE +/- 19.00, N = 61954.381966.911967.161998.442019.671. (CXX) g++ options: -O3 -lsnappy -lpthread
OpenBenchmarking.orgMicroseconds Per Op, Fewer Is BetterLevelDB 1.22Benchmark: Fill SyncNoneMQ-DeadlineKyberBFQ low-latencyBFQ400800120016002000Min: 1947.65 / Avg: 1954.38 / Max: 1961.23Min: 1959.76 / Avg: 1966.91 / Max: 1978.13Min: 1942.11 / Avg: 1967.16 / Max: 1988.21Min: 1972.85 / Avg: 1998.44 / Max: 2036.25Min: 1963.38 / Avg: 2019.67 / Max: 2067.391. (CXX) g++ options: -O3 -lsnappy -lpthread

OpenBenchmarking.orgMB/s, More Is BetterLevelDB 1.22Benchmark: Fill SyncKyberMQ-DeadlineNoneBFQBFQ low-latency0.811.622.433.244.05SE +/- 0.03, N = 3SE +/- 0.03, N = 3SE +/- 0.00, N = 3SE +/- 0.04, N = 6SE +/- 0.06, N = 33.63.63.63.53.51. (CXX) g++ options: -O3 -lsnappy -lpthread
OpenBenchmarking.orgMB/s, More Is BetterLevelDB 1.22Benchmark: Fill SyncKyberMQ-DeadlineNoneBFQBFQ low-latency246810Min: 3.5 / Avg: 3.57 / Max: 3.6Min: 3.5 / Avg: 3.57 / Max: 3.6Min: 3.6 / Avg: 3.6 / Max: 3.6Min: 3.4 / Avg: 3.48 / Max: 3.6Min: 3.4 / Avg: 3.5 / Max: 3.61. (CXX) g++ options: -O3 -lsnappy -lpthread

IOR

IOR is a parallel I/O storage benchmark. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterIOR 3.2.1Read TestNoneBFQBFQ low-latencyKyberMQ-Deadline2004006008001000SE +/- 0.64, N = 3SE +/- 2.94, N = 3SE +/- 5.02, N = 3SE +/- 5.68, N = 3SE +/- 1.94, N = 3995.19986.32978.41977.87977.23MIN: 727.61 / MAX: 1323.38MIN: 715.91 / MAX: 1357.45MIN: 724.27 / MAX: 1310.08MIN: 718.27 / MAX: 1337.12MIN: 746.57 / MAX: 1252.841. (CC) gcc options: -O2 -lm -pthread -lmpi
OpenBenchmarking.orgMB/s, More Is BetterIOR 3.2.1Read TestNoneBFQBFQ low-latencyKyberMQ-Deadline2004006008001000Min: 994.16 / Avg: 995.19 / Max: 996.37Min: 980.69 / Avg: 986.32 / Max: 990.57Min: 972.67 / Avg: 978.41 / Max: 988.42Min: 966.52 / Avg: 977.87 / Max: 983.89Min: 974.12 / Avg: 977.23 / Max: 980.81. (CC) gcc options: -O2 -lm -pthread -lmpi

OpenBenchmarking.orgMB/s, More Is BetterIOR 3.2.1Write TestBFQ low-latencyNoneMQ-DeadlineKyberBFQ2004006008001000SE +/- 4.45, N = 3SE +/- 3.67, N = 3SE +/- 6.47, N = 3SE +/- 5.10, N = 3SE +/- 5.04, N = 3799.22797.73794.35781.43770.48MIN: 156.94 / MAX: 903.57MIN: 165.18 / MAX: 906.83MIN: 162.92 / MAX: 909.98MIN: 13.98 / MAX: 900.79MIN: 13.92 / MAX: 898.051. (CC) gcc options: -O2 -lm -pthread -lmpi
OpenBenchmarking.orgMB/s, More Is BetterIOR 3.2.1Write TestBFQ low-latencyNoneMQ-DeadlineKyberBFQ140280420560700Min: 793.69 / Avg: 799.22 / Max: 808.02Min: 790.51 / Avg: 797.73 / Max: 802.48Min: 781.47 / Avg: 794.35 / Max: 801.91Min: 771.59 / Avg: 781.43 / Max: 788.66Min: 760.81 / Avg: 770.48 / Max: 777.81. (CC) gcc options: -O2 -lm -pthread -lmpi

52 Results Shown

MariaDB
Apache Cassandra
Timed GCC Compilation
SQLite
Apache HBase:
  Async Rand Write - 32:
    Microseconds - Average Latency
    Rows Per Second
PostgreSQL pgbench
Apache Cassandra
Facebook RocksDB
Apache HBase:
  Seq Read - 32:
    Microseconds - Average Latency
    Rows Per Second
Flexible IO Tester:
  Seq Write - IO_uring - Yes - No - 2MB - Default Test Directory:
    IOPS
    MB/s
LevelDB:
  Rand Delete
  Seq Fill
  Seq Fill
FS-Mark
Flexible IO Tester:
  Rand Write - IO_uring - Yes - No - 4KB - Default Test Directory:
    IOPS
    MB/s
Timed GDB GNU Debugger Compilation
Facebook RocksDB
KeyDB
Facebook RocksDB
LevelDB
Apache HBase:
  Increment - 32:
    Microseconds - Average Latency
    Rows Per Second
Facebook RocksDB:
  Rand Fill
  Seq Fill
Flexible IO Tester:
  Seq Write - IO_uring - Yes - No - 4KB - Default Test Directory:
    IOPS
    MB/s
  Rand Read - IO_uring - Yes - No - 4KB - Default Test Directory:
    IOPS
    MB/s
LevelDB
Apache HBase:
  Async Rand Read - 32:
    Microseconds - Average Latency
    Rows Per Second
LevelDB
Flexible IO Tester:
  Seq Read - IO_uring - Yes - No - 4KB - Default Test Directory:
    IOPS
    MB/s
SQLite
Flexible IO Tester:
  Seq Read - IO_uring - Yes - No - 2MB - Default Test Directory:
    IOPS
    MB/s
Apache HBase:
  Rand Read - 32:
    Microseconds - Average Latency
    Rows Per Second
LevelDB:
  Rand Fill:
    Microseconds Per Op
    MB/s
  Overwrite:
    Microseconds Per Op
    MB/s
FS-Mark
LevelDB:
  Fill Sync:
    Microseconds Per Op
    MB/s
IOR:
  Read Test
  Write Test

None

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: NONE / errors=remount-ro,relatime,rw
Processor Notes: Scaling Governor: acpi-cpufreq ondemand - CPU Microcode: 0x8301025
Java Notes: OpenJDK Runtime Environment (build 11.0.6+10-post-Ubuntu-2ubuntu2)
Security Notes: itlb_multihit: Not affected + 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 Full AMD retpoline IBPB: conditional STIBP: conditional RSB filling + tsx_async_abort: Not affected

Testing initiated at 23 March 2020 08:04 by user phoronix.

MQ-Deadline

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: acpi-cpufreq ondemand - CPU Microcode: 0x8301025
Java Notes: OpenJDK Runtime Environment (build 11.0.6+10-post-Ubuntu-2ubuntu2)
Security Notes: itlb_multihit: Not affected + 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 Full AMD retpoline IBPB: conditional STIBP: conditional RSB filling + tsx_async_abort: Not affected

Testing initiated at 23 March 2020 12:43 by user phoronix.

Kyber

Processor: AMD Ryzen Threadripper 3970X 32-Core @ 3.70GHz (32 Cores / 64 Threads), Motherboard: ASUS ROG ZENITH II EXTREME (0702 BIOS), Chipset: AMD Starship/Matisse, Memory: 64GB, Disk: 1000GB Force MP600, Graphics: Gigabyte AMD Radeon RX 5500/5500M / Pro 5500M 8GB (1890/875MHz), Audio: AMD Navi 10 HDMI Audio, Monitor: ASUS MG28U, Network: Aquantia AQC107 NBase-T/IEEE + Intel I211 + Intel Wi-Fi 6 AX200

OS: Ubuntu 20.04, Kernel: 5.6.0-050600rc7-generic (x86_64) 20200322, Desktop: GNOME Shell 3.35.91, Display Server: X Server 1.20.7, Display Driver: modesetting 1.20.7, OpenGL: 4.6 Mesa 20.0.0 (LLVM 9.0.1), Compiler: GCC 9.3.0, File-System: ext4, Screen Resolution: 1280x800

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: KYBER / errors=remount-ro,relatime,rw
Processor Notes: Scaling Governor: acpi-cpufreq ondemand - CPU Microcode: 0x8301025
Java Notes: OpenJDK Runtime Environment (build 11.0.6+10-post-Ubuntu-2ubuntu2)
Security Notes: itlb_multihit: Not affected + 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 Full AMD retpoline IBPB: conditional STIBP: conditional RSB filling + tsx_async_abort: Not affected

Testing initiated at 24 March 2020 05:42 by user phoronix.

BFQ low-latency

Processor: AMD Ryzen Threadripper 3970X 32-Core @ 3.70GHz (32 Cores / 64 Threads), Motherboard: ASUS ROG ZENITH II EXTREME (0702 BIOS), Chipset: AMD Starship/Matisse, Memory: 64GB, Disk: 1000GB Force MP600, Graphics: Gigabyte AMD Radeon RX 5500/5500M / Pro 5500M 8GB (1890/875MHz), Audio: AMD Navi 10 HDMI Audio, Monitor: ASUS MG28U, Network: Aquantia AQC107 NBase-T/IEEE + Intel I211 + Intel Wi-Fi 6 AX200

OS: Ubuntu 20.04, Kernel: 5.6.0-050600rc7-generic (x86_64) 20200322, Desktop: GNOME Shell 3.35.91, Display Server: X Server 1.20.7, Display Driver: modesetting 1.20.7, OpenGL: 4.6 Mesa 20.0.0 (LLVM 9.0.1), Compiler: GCC 9.3.0, File-System: ext4, Screen Resolution: 3840x2160

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: BFQ / errors=remount-ro,relatime,rw
Processor Notes: Scaling Governor: acpi-cpufreq ondemand - CPU Microcode: 0x8301025
Java Notes: OpenJDK Runtime Environment (build 11.0.6+10-post-Ubuntu-2ubuntu2)
Security Notes: itlb_multihit: Not affected + 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 Full AMD retpoline IBPB: conditional STIBP: conditional RSB filling + tsx_async_abort: Not affected

Testing initiated at 24 March 2020 20:05 by user phoronix.

BFQ

Processor: AMD Ryzen Threadripper 3970X 32-Core @ 3.70GHz (32 Cores / 64 Threads), Motherboard: ASUS ROG ZENITH II EXTREME (0702 BIOS), Chipset: AMD Starship/Matisse, Memory: 64GB, Disk: 1000GB Force MP600, Graphics: Gigabyte AMD Radeon RX 5500/5500M / Pro 5500M 8GB (1890/875MHz), Audio: AMD Navi 10 HDMI Audio, Monitor: ASUS MG28U, Network: Aquantia AQC107 NBase-T/IEEE + Intel I211 + Intel Wi-Fi 6 AX200

OS: Ubuntu 20.04, Kernel: 5.6.0-050600rc7-generic (x86_64) 20200322, Desktop: GNOME Shell 3.35.91, Display Server: X Server 1.20.7, Display Driver: modesetting 1.20.7, OpenGL: 4.6 Mesa 20.0.0 (LLVM 9.0.1), Compiler: GCC 9.3.0, File-System: ext4, Screen Resolution: 1280x800

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: BFQ / errors=remount-ro,relatime,rw
Processor Notes: Scaling Governor: acpi-cpufreq ondemand - CPU Microcode: 0x8301025
Java Notes: OpenJDK Runtime Environment (build 11.0.6+10-post-Ubuntu-2ubuntu2)
Security Notes: itlb_multihit: Not affected + 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 Full AMD retpoline IBPB: conditional STIBP: conditional RSB filling + tsx_async_abort: Not affected

Testing initiated at 25 March 2020 09:59 by user phoronix.