HP DL380p Gen8 4 disk RAID5 vs RAID1+0

Benchmarking the difference between RAID 5 and RAID 1+0 on a HP ProLiant DL380p Gen8, using a P420i (revision B, firmware 6.00) card with 1GB cache (90% write cache), strip size of 256KB and 4 x 1.2TB 10K SFF SAS hard drives. The OS, a fresh CentOS 6.6 installation, was installed on a separate RAID1 array.

Compare your own system(s) to this result file with the Phoronix Test Suite by running the command: phoronix-test-suite benchmark 1411187-REZZ-HPDL38097
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:

C/C++ Compiler Tests 2 Tests
CPU Massive 3 Tests
Database Test Suite 2 Tests
Disk Test Suite 6 Tests
Common Kernel Benchmarks 3 Tests
Server 3 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
Performance Per
Dollar
Date
Run
  Test
  Duration
RAID5
November 17 2014
 
RAID1+0
November 18 2014
 
Invert Hiding All Results Option
 
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):

RAID5

Compiler Notes: --build=x86_64-redhat-linux --disable-dssi --disable-libjava-multilib --disable-libunwind-exceptions --enable-__cxa_atexit --enable-bootstrap --enable-checking=release --enable-gnu-unique-object --enable-java-awt=gtk --enable-java-maintainer-mode --enable-languages=c,c++,objc,obj-c++,java,fortran,ada --enable-libgcj-multifile --enable-shared --enable-threads=posix --mandir=/usr/share/man --with-arch_32=i686 --with-cloog --with-ppl --with-tune=generic
Disk Notes: CFQ / barrier=1,data=ordered,relatime,rw,stripe=64
Processor Notes: Scaling Governor: pcc-cpufreq userspace
System Notes: Disk Scheduler: CFQ. Python 2.6.6.

RAID1+0

Processor: Intel Xeon E5-2620 v2 @ 2.10GHz (12 Cores), Motherboard: HP ProLiant DL380p Gen8, Chipset: Intel Xeon E7 v2/Xeon, Memory: 16384MB, Disk: 300GB LOGICAL VOLUME + 3601GB LOGICAL VOLUME, Graphics: Matrox s MGA G200EH, Monitor: Smart Cable, Network: Broadcom NetXtreme BCM5719 Gigabit PCIe

OS: CentOS 6.6, Kernel: 2.6.32-504.1.3.el6.x86_64 (x86_64), Display Driver: modesetting 0.8.0, Compiler: GCC 4.4.7 20120313, File-System: ext4

HP DL380p Gen8 4 disk RAID5 vs RAID1+0OpenBenchmarking.orgPhoronix Test SuiteIntel Xeon E5-2620 v2 @ 2.10GHz (12 Cores)HP ProLiant DL380p Gen8Intel Xeon E7 v2/Xeon16384MB300GB LOGICAL VOLUME + 3601GB LOGICAL VOLUMEMatrox s MGA G200EHSmart CableBroadcom NetXtreme BCM5719 Gigabit PCIeCentOS 6.62.6.32-504.1.3.el6.x86_64 (x86_64)modesetting 0.8.0GCC 4.4.7 20120313ext4ProcessorMotherboardChipsetMemoryDiskGraphicsMonitorNetworkOSKernelDisplay DriverCompilerFile-SystemHP DL380p Gen8 4 Disk RAID5 Vs RAID1+0 BenchmarksSystem Logs- --build=x86_64-redhat-linux --disable-dssi --disable-libjava-multilib --disable-libunwind-exceptions --enable-__cxa_atexit --enable-bootstrap --enable-checking=release --enable-gnu-unique-object --enable-java-awt=gtk --enable-java-maintainer-mode --enable-languages=c,c++,objc,obj-c++,java,fortran,ada --enable-libgcj-multifile --enable-shared --enable-threads=posix --mandir=/usr/share/man --with-arch_32=i686 --with-cloog --with-ppl --with-tune=generic - CFQ / barrier=1,data=ordered,relatime,rw,stripe=64- Scaling Governor: pcc-cpufreq userspace- Disk Scheduler: CFQ. Python 2.6.6.

RAID5 vs. RAID1+0 ComparisonPhoronix Test SuiteBaseline+68.9%+68.9%+137.8%+137.8%+206.7%+206.7%275.5%5.5%3.8%3.5%2.2%64MB Rand Write - 32 Threads8.W.P44.7%64MB Rand Read - 32 Threads8.2%Compile6.5%T.B.T.P.SInitial Create1.F.1.S12,500 INSERTsThreaded I/O TesterIOzoneThreaded I/O TesterCompile BenchPostgreSQL pgbenchCompile BenchFS-MarkSQLiteRAID5RAID1+0

HP DL380p Gen8 4 disk RAID5 vs RAID1+0aio-stress: Rand Writesqlite: 12,500 INSERTsfs-mark: 1000 Files, 1MB Sizedbench: 12 Clientsdbench: 48 Clientsdbench: 128 Clientsdbench: 1 Clientsiozone: 8GB Read Performanceiozone: 8GB Write Performancetiobench: 64MB Rand Read - 32 Threadstiobench: 64MB Rand Write - 32 Threadscompilebench: Compilecompilebench: Initial Createcompilebench: Read Compiled Treeunpack-linux: linux-2.6.32.tar.bz2postmark: Disk Transaction Performancecompress-gzip: 2GB File Compressionpgbench: TPC-B Transactions Per Secondapache: Static Web Page ServingRAID5RAID1+02422.776.60637.682481.363207.282909.55605.906252.36545.226044.46100.82774.83181.401216.1612.02375017.916993.4617539.742402.776.46659.802495.303217.172920.24606.336284.78376.695588.19378.58727.67188.381194.6012.04375017.927377.0517420.15OpenBenchmarking.org

AIO-Stress

AIO-Stress is an a-synchronous I/O benchmark created by SuSE. Current this profile uses a 2048MB test file and a 64KB record size. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterAIO-Stress 0.21Random WriteRAID5RAID1+05001000150020002500SE +/- 7.68, N = 3SE +/- 2.11, N = 32422.772402.771. (CC) gcc options: -pthread -laio
OpenBenchmarking.orgMB/s, More Is BetterAIO-Stress 0.21Random WriteRAID5RAID1+0400800120016002000Min: 2413.78 / Avg: 2422.77 / Max: 2438.05Min: 2398.6 / Avg: 2402.77 / Max: 2405.371. (CC) gcc options: -pthread -laio

SQLite

OpenBenchmarking.orgSeconds, Fewer Is BetterSQLite 3.7.312,500 INSERTsRAID5RAID1+0246810SE +/- 0.10, N = 5SE +/- 0.02, N = 36.606.461. (CC) gcc options: -O2 -lreadline -lcurses -ldl -lpthread
OpenBenchmarking.orgSeconds, Fewer Is BetterSQLite 3.7.312,500 INSERTsRAID5RAID1+03691215Min: 6.46 / Avg: 6.6 / Max: 7Min: 6.42 / Avg: 6.46 / Max: 6.51. (CC) gcc options: -O2 -lreadline -lcurses -ldl -lpthread

FS-Mark

OpenBenchmarking.orgFiles/s, More Is BetterFS-Mark 3.31000 Files, 1MB SizeRAID5RAID1+0140280420560700SE +/- 23.26, N = 6SE +/- 23.26, N = 6637.68659.801. (CC) gcc options: -static
OpenBenchmarking.orgFiles/s, More Is BetterFS-Mark 3.31000 Files, 1MB SizeRAID5RAID1+0120240360480600Min: 544.9 / Avg: 637.68 / Max: 681.7Min: 544.7 / Avg: 659.8 / Max: 692.71. (CC) gcc options: -static

Dbench

Dbench is a benchmark designed by the Samba project as a free alternative to netbench, but dbench contains only file-system calls for testing the disk performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterDbench 4.012 ClientsRAID5RAID1+05001000150020002500SE +/- 2.43, N = 3SE +/- 1.09, N = 32481.362495.301. (CC) gcc options: -lpopt -O2
OpenBenchmarking.orgMB/s, More Is BetterDbench 4.012 ClientsRAID5RAID1+0400800120016002000Min: 2476.72 / Avg: 2481.36 / Max: 2484.95Min: 2493.11 / Avg: 2495.3 / Max: 2496.441. (CC) gcc options: -lpopt -O2

OpenBenchmarking.orgMB/s, More Is BetterDbench 4.048 ClientsRAID5RAID1+07001400210028003500SE +/- 3.31, N = 3SE +/- 0.73, N = 33207.283217.171. (CC) gcc options: -lpopt -O2
OpenBenchmarking.orgMB/s, More Is BetterDbench 4.048 ClientsRAID5RAID1+06001200180024003000Min: 3201.16 / Avg: 3207.28 / Max: 3212.54Min: 3215.83 / Avg: 3217.17 / Max: 3218.331. (CC) gcc options: -lpopt -O2

OpenBenchmarking.orgMB/s, More Is BetterDbench 4.0128 ClientsRAID5RAID1+06001200180024003000SE +/- 5.02, N = 3SE +/- 18.78, N = 32909.552920.241. (CC) gcc options: -lpopt -O2
OpenBenchmarking.orgMB/s, More Is BetterDbench 4.0128 ClientsRAID5RAID1+05001000150020002500Min: 2902.69 / Avg: 2909.55 / Max: 2919.33Min: 2883.52 / Avg: 2920.24 / Max: 2945.41. (CC) gcc options: -lpopt -O2

OpenBenchmarking.orgMB/s, More Is BetterDbench 4.01 ClientsRAID5RAID1+0130260390520650SE +/- 2.89, N = 3SE +/- 2.10, N = 3605.90606.331. (CC) gcc options: -lpopt -O2
OpenBenchmarking.orgMB/s, More Is BetterDbench 4.01 ClientsRAID5RAID1+0110220330440550Min: 600.11 / Avg: 605.9 / Max: 608.9Min: 603.44 / Avg: 606.33 / Max: 610.41. (CC) gcc options: -lpopt -O2

IOzone

The IOzone benchmark tests the hard disk drive / file-system performance. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterIOzone 3.4058GB Read PerformanceRAID5RAID1+013002600390052006500SE +/- 65.15, N = 3SE +/- 15.57, N = 36252.366284.781. (CC) gcc options: -O3
OpenBenchmarking.orgMB/s, More Is BetterIOzone 3.4058GB Read PerformanceRAID5RAID1+011002200330044005500Min: 6122.13 / Avg: 6252.36 / Max: 6321.39Min: 6254.49 / Avg: 6284.78 / Max: 6306.221. (CC) gcc options: -O3

OpenBenchmarking.orgMB/s, More Is BetterIOzone 3.4058GB Write PerformanceRAID5RAID1+0120240360480600SE +/- 1.04, N = 3SE +/- 0.86, N = 3545.22376.691. (CC) gcc options: -O3
OpenBenchmarking.orgMB/s, More Is BetterIOzone 3.4058GB Write PerformanceRAID5RAID1+0100200300400500Min: 543.52 / Avg: 545.22 / Max: 547.12Min: 375.05 / Avg: 376.69 / Max: 377.961. (CC) gcc options: -O3

Threaded I/O Tester

OpenBenchmarking.orgMB/s, More Is BetterThreaded I/O Tester 0.3.364MB Random Read - 32 ThreadsRAID5RAID1+013002600390052006500SE +/- 47.49, N = 3SE +/- 5.82, N = 36044.465588.19
OpenBenchmarking.orgMB/s, More Is BetterThreaded I/O Tester 0.3.364MB Random Read - 32 ThreadsRAID5RAID1+010002000300040005000Min: 5950.4 / Avg: 6044.46 / Max: 6102.92Min: 5577.12 / Avg: 5588.19 / Max: 5596.85

OpenBenchmarking.orgMB/s, More Is BetterThreaded I/O Tester 0.3.364MB Random Write - 32 ThreadsRAID5RAID1+080160240320400SE +/- 60.32, N = 6SE +/- 4.28, N = 3100.82378.58
OpenBenchmarking.orgMB/s, More Is BetterThreaded I/O Tester 0.3.364MB Random Write - 32 ThreadsRAID5RAID1+070140210280350Min: 11.43 / Avg: 100.82 / Max: 391.67Min: 372.97 / Avg: 378.58 / Max: 386.97

Compile Bench

Compilebench tries to age a filesystem by simulating some of the disk IO common in creating, compiling, patching, stating and reading kernel trees. It indirectly measures how well filesystems can maintain directory locality as the disk fills up and directories age. This current test is setup to use the makej mode with 10 initial directories Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgMB/s, More Is BetterCompile Bench 0.6Test: CompileRAID5RAID1+02004006008001000SE +/- 42.16, N = 6SE +/- 14.64, N = 6774.83727.67
OpenBenchmarking.orgMB/s, More Is BetterCompile Bench 0.6Test: CompileRAID5RAID1+0140280420560700Min: 566.5 / Avg: 774.83 / Max: 834.72Min: 657.35 / Avg: 727.67 / Max: 753.57

OpenBenchmarking.orgMB/s, More Is BetterCompile Bench 0.6Test: Initial CreateRAID5RAID1+04080120160200SE +/- 7.52, N = 3SE +/- 2.32, N = 3181.40188.38
OpenBenchmarking.orgMB/s, More Is BetterCompile Bench 0.6Test: Initial CreateRAID5RAID1+0306090120150Min: 166.58 / Avg: 181.4 / Max: 191.01Min: 183.88 / Avg: 188.38 / Max: 191.63

OpenBenchmarking.orgMB/s, More Is BetterCompile Bench 0.6Test: Read Compiled TreeRAID5RAID1+030060090012001500SE +/- 1.65, N = 3SE +/- 3.67, N = 31216.161194.60
OpenBenchmarking.orgMB/s, More Is BetterCompile Bench 0.6Test: Read Compiled TreeRAID5RAID1+02004006008001000Min: 1214.45 / Avg: 1216.16 / Max: 1219.46Min: 1187.27 / Avg: 1194.6 / Max: 1198.27

Unpacking The Linux Kernel

OpenBenchmarking.orgSeconds, Fewer Is BetterUnpacking The Linux Kernellinux-2.6.32.tar.bz2RAID5RAID1+03691215SE +/- 0.20, N = 4SE +/- 0.20, N = 412.0212.04
OpenBenchmarking.orgSeconds, Fewer Is BetterUnpacking The Linux Kernellinux-2.6.32.tar.bz2RAID5RAID1+048121620Min: 11.81 / Avg: 12.02 / Max: 12.61Min: 11.83 / Avg: 12.04 / Max: 12.65

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 PerformanceRAID5RAID1+08001600240032004000SE +/- 18.67, N = 3SE +/- 18.67, N = 3375037501. (CC) gcc options: -O3
OpenBenchmarking.orgTPS, More Is BetterPostMark 1.51Disk Transaction PerformanceRAID5RAID1+07001400210028003500Min: 3731 / Avg: 3749.67 / Max: 3787Min: 3731 / Avg: 3749.67 / Max: 37871. (CC) gcc options: -O3

Gzip Compression

OpenBenchmarking.orgSeconds, Fewer Is BetterGzip Compression2GB File CompressionRAID5RAID1+048121620SE +/- 0.10, N = 3SE +/- 0.11, N = 317.9117.92
OpenBenchmarking.orgSeconds, Fewer Is BetterGzip Compression2GB File CompressionRAID5RAID1+0510152025Min: 17.71 / Avg: 17.91 / Max: 18Min: 17.71 / Avg: 17.92 / Max: 18.06

PostgreSQL pgbench

OpenBenchmarking.orgTPS, More Is BetterPostgreSQL pgbench 8.4.11TPC-B Transactions Per SecondRAID5RAID1+016003200480064008000SE +/- 105.20, N = 6SE +/- 18.37, N = 36993.467377.051. (CC) gcc options: -O2 -fno-strict-aliasing -fwrapv -lpgport -lpq -lcrypt -ldl -lm
OpenBenchmarking.orgTPS, More Is BetterPostgreSQL pgbench 8.4.11TPC-B Transactions Per SecondRAID5RAID1+013002600390052006500Min: 6512.32 / Avg: 6993.46 / Max: 7223.09Min: 7350.9 / Avg: 7377.05 / Max: 7412.491. (CC) gcc options: -O2 -fno-strict-aliasing -fwrapv -lpgport -lpq -lcrypt -ldl -lm

Apache Benchmark

This is a test of ab, which is the Apache benchmark program. This test profile measures how many requests per second a given system can sustain when carrying out 1,000,000 requests with 100 requests being carried out concurrently. Learn more via the OpenBenchmarking.org test page.

OpenBenchmarking.orgRequests Per Second, More Is BetterApache Benchmark 2.4.7Static Web Page ServingRAID5RAID1+04K8K12K16K20KSE +/- 50.73, N = 3SE +/- 102.33, N = 317539.7417420.151. (CC) gcc options: -shared -fPIC -O2 -pthread
OpenBenchmarking.orgRequests Per Second, More Is BetterApache Benchmark 2.4.7Static Web Page ServingRAID5RAID1+03K6K9K12K15KMin: 17474.16 / Avg: 17539.74 / Max: 17639.57Min: 17271.45 / Avg: 17420.15 / Max: 17616.271. (CC) gcc options: -shared -fPIC -O2 -pthread

RAID5

Compiler Notes: --build=x86_64-redhat-linux --disable-dssi --disable-libjava-multilib --disable-libunwind-exceptions --enable-__cxa_atexit --enable-bootstrap --enable-checking=release --enable-gnu-unique-object --enable-java-awt=gtk --enable-java-maintainer-mode --enable-languages=c,c++,objc,obj-c++,java,fortran,ada --enable-libgcj-multifile --enable-shared --enable-threads=posix --mandir=/usr/share/man --with-arch_32=i686 --with-cloog --with-ppl --with-tune=generic
Disk Notes: CFQ / barrier=1,data=ordered,relatime,rw,stripe=64
Processor Notes: Scaling Governor: pcc-cpufreq userspace
System Notes: Disk Scheduler: CFQ. Python 2.6.6.

Testing initiated at 17 November 2014 22:19 by user .

RAID1+0

Processor: Intel Xeon E5-2620 v2 @ 2.10GHz (12 Cores), Motherboard: HP ProLiant DL380p Gen8, Chipset: Intel Xeon E7 v2/Xeon, Memory: 16384MB, Disk: 300GB LOGICAL VOLUME + 3601GB LOGICAL VOLUME, Graphics: Matrox s MGA G200EH, Monitor: Smart Cable, Network: Broadcom NetXtreme BCM5719 Gigabit PCIe

OS: CentOS 6.6, Kernel: 2.6.32-504.1.3.el6.x86_64 (x86_64), Display Driver: modesetting 0.8.0, Compiler: GCC 4.4.7 20120313, File-System: ext4

Compiler Notes: --build=x86_64-redhat-linux --disable-dssi --disable-libjava-multilib --disable-libunwind-exceptions --enable-__cxa_atexit --enable-bootstrap --enable-checking=release --enable-gnu-unique-object --enable-java-awt=gtk --enable-java-maintainer-mode --enable-languages=c,c++,objc,obj-c++,java,fortran,ada --enable-libgcj-multifile --enable-shared --enable-threads=posix --mandir=/usr/share/man --with-arch_32=i686 --with-cloog --with-ppl --with-tune=generic
Disk Notes: CFQ / barrier=1,data=ordered,relatime,rw,stripe=64
Processor Notes: Scaling Governor: pcc-cpufreq userspace
System Notes: Disk Scheduler: CFQ. Python 2.6.6.

Testing initiated at 18 November 2014 09:58 by user .