1Gbpsの高速なネットワークをゆったり利用できるNTTPCコミュニケーションズのVPSサービスを試す 5ページ
付録：UnixBenchおよびFS-Mark、Threaded I/O testerによるベンチマークテスト結果の詳細
リストA UnixBenchの実行結果
------------------------------------------------------------------------
========================================================================
BYTE UNIX Benchmarks (Version 5.1.3)
System: localhost.localdomain: GNU/Linux
OS: GNU/Linux -- 2.6.18-238.5.1.el5.028stab085.5.nttpc -- #1 SMP Tue Apr 26 19:53:50 JST 2011
Machine: x86_64 (x86_64)
Language: en_US.utf8 (charmap="UTF-8", collate="UTF-8")
CPU 0: Intel(R) Xeon(R) CPU X5650 @ 2.67GHz (5333.7 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET, Intel virtualization
18:18:28 up 9 days, 16:00, 1 user, load average: 0.08, 0.19, 0.11; runlevel
------------------------------------------------------------------------
Benchmark Run: Thu Feb 16 2012 18:18:28 - 18:46:21
1 CPU in system; running 1 parallel copy of tests
Dhrystone 2 using register variables 13431520.3 lps (10.0 s, 7 samples)
Double-Precision Whetstone 2862.8 MWIPS (8.0 s, 7 samples)
Execl Throughput 3621.7 lps (29.9 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 728904.2 KBps (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 208806.3 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 1356077.2 KBps (30.0 s, 2 samples)
Pipe Throughput 1876789.2 lps (10.0 s, 7 samples)
Pipe-based Context Switching 464289.0 lps (10.0 s, 7 samples)
Process Creation 11937.8 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 4788.9 lpm (60.0 s, 2 samples)
Shell Scripts (8 concurrent) 684.0 lpm (60.0 s, 2 samples)
System Call Overhead 3647438.9 lps (10.0 s, 7 samples)
System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 13431520.3 1150.9
Double-Precision Whetstone 55.0 2862.8 520.5
Execl Throughput 43.0 3621.7 842.3
File Copy 1024 bufsize 2000 maxblocks 3960.0 728904.2 1840.7
File Copy 256 bufsize 500 maxblocks 1655.0 208806.3 1261.7
File Copy 4096 bufsize 8000 maxblocks 5800.0 1356077.2 2338.1
Pipe Throughput 12440.0 1876789.2 1508.7
Pipe-based Context Switching 4000.0 464289.0 1160.7
Process Creation 126.0 11937.8 947.4
Shell Scripts (1 concurrent) 42.4 4788.9 1129.5
Shell Scripts (8 concurrent) 6.0 684.0 1140.0
System Call Overhead 15000.0 3647438.9 2431.6
========
System Benchmarks Index Score 1247.7
リストB FS-Markの実行結果全文
$ ./fs_mark -d ./tmp -s 1048576 -n 1000
# ./fs_mark -d ./tmp -s 1048576 -n 1000
# Version 3.3, 1 thread(s) starting at Thu Feb 16 18:49:28 2012
# Sync method: INBAND FSYNC: fsync() per file in write loop.
# Directories: no subdirectories used
# File names: 40 bytes long, (16 initial bytes of time stamp with 24 random bytes at end of name)
# Files info: size 1048576 bytes, written with an IO size of 16384 bytes per write
# App overhead is time in microseconds spent in the test not doing file writing related system calls.
FSUse% Count Size Files/sec App Overhead
2 1000 1048576 110.9 22561
$ ./fs_mark -d ./tmp -s 1048576 -n 1000
# ./fs_mark -d ./tmp -s 1048576 -n 1000
# Version 3.3, 1 thread(s) starting at Thu Feb 16 18:50:02 2012
# Sync method: INBAND FSYNC: fsync() per file in write loop.
# Directories: no subdirectories used
# File names: 40 bytes long, (16 initial bytes of time stamp with 24 random bytes at end of name)
# Files info: size 1048576 bytes, written with an IO size of 16384 bytes per write
# App overhead is time in microseconds spent in the test not doing file writing related system calls.
FSUse% Count Size Files/sec App Overhead
2 1000 1048576 199.6 21245
42
$ ./fs_mark -d ./tmp -s 1048576 -n 1000
# ./fs_mark -d ./tmp -s 1048576 -n 1000
# Version 3.3, 1 thread(s) starting at Thu Feb 16 19:17:53 2012
# Sync method: INBAND FSYNC: fsync() per file in write loop.
# Directories: no subdirectories used
# File names: 40 bytes long, (16 initial bytes of time stamp with 24 random bytes at end of name)
# Files info: size 1048576 bytes, written with an IO size of 16384 bytes per write
# App overhead is time in microseconds spent in the test not doing file writing related system calls.
FSUse% Count Size Files/sec App Overhead
2 1000 1048576 91.9 22589
リストC Threaded I/O testerの実行結果全文
$ ./tiotest -f 32 -t 32 Tiotest results for 32 concurrent io threads: ,----------------------------------------------------------------------. | Item | Time | Rate | Usr CPU | Sys CPU | +-----------------------+----------+--------------+----------+---------+ | Write 1024 MBs | 4.4 s | 234.023 MB/s | 9.9 % | 357.8 % | | Random Write 125 MBs | 0.3 s | 445.661 MB/s | 36.0 % | 404.2 % | | Read 1024 MBs | 0.5 s | 2217.377 MB/s | 10.0 % | 89.4 % | | Random Read 125 MBs | 0.0 s | 2989.072 MB/s | 12.0 % | 83.7 % | `----------------------------------------------------------------------' Tiotest latency results: ,-------------------------------------------------------------------------. | Item | Average latency | Maximum latency | % >2 sec | % >10 sec | +--------------+-----------------+-----------------+----------+-----------+ | Write | 0.022 ms | 300.073 ms | 0.00000 | 0.00000 | | Random Write | 0.006 ms | 57.930 ms | 0.00000 | 0.00000 | | Read | 0.002 ms | 1.617 ms | 0.00000 | 0.00000 | | Random Read | 0.001 ms | 0.079 ms | 0.00000 | 0.00000 | |--------------+-----------------+-----------------+----------+-----------| | Total | 0.011 ms | 300.073 ms | 0.00000 | 0.00000 | `--------------+-----------------+-----------------+----------+-----------' $ ./tiotest -f 32 -t 32 Tiotest results for 32 concurrent io threads: ,----------------------------------------------------------------------. | Item | Time | Rate | Usr CPU | Sys CPU | +-----------------------+----------+--------------+----------+---------+ | Write 1024 MBs | 3.7 s | 277.413 MB/s | 17.3 % | 599.0 % | | Random Write 125 MBs | 1.1 s | 113.486 MB/s | 7.3 % | 169.6 % | | Read 1024 MBs | 0.4 s | 2647.377 MB/s | 13.4 % | 85.6 % | | Random Read 125 MBs | 0.0 s | 2677.692 MB/s | 25.7 % | 72.8 % | `----------------------------------------------------------------------' Tiotest latency results: ,-------------------------------------------------------------------------. | Item | Average latency | Maximum latency | % >2 sec | % >10 sec | +--------------+-----------------+-----------------+----------+-----------+ | Write | 0.052 ms | 1910.076 ms | 0.00000 | 0.00000 | | Random Write | 0.002 ms | 0.024 ms | 0.00000 | 0.00000 | | Read | 0.001 ms | 0.169 ms | 0.00000 | 0.00000 | | Random Read | 0.001 ms | 0.185 ms | 0.00000 | 0.00000 | |--------------+-----------------+-----------------+----------+-----------| | Total | 0.024 ms | 1910.076 ms | 0.00000 | 0.00000 | `--------------+-----------------+-----------------+----------+-----------' $ ./tiotest -f 32 -t 32 Tiotest results for 32 concurrent io threads: ,----------------------------------------------------------------------. | Item | Time | Rate | Usr CPU | Sys CPU | +-----------------------+----------+--------------+----------+---------+ | Write 1024 MBs | 5.1 s | 200.290 MB/s | 19.4 % | 760.7 % | | Random Write 125 MBs | 0.3 s | 427.236 MB/s | 69.4 % | 620.3 % | | Read 1024 MBs | 0.4 s | 2728.782 MB/s | 13.1 % | 85.5 % | | Random Read 125 MBs | 0.0 s | 2771.926 MB/s | 15.5 % | 79.8 % | `----------------------------------------------------------------------' Tiotest latency results: ,-------------------------------------------------------------------------. | Item | Average latency | Maximum latency | % >2 sec | % >10 sec | +--------------+-----------------+-----------------+----------+-----------+ | Write | 0.314 ms | 2558.833 ms | 0.00076 | 0.00000 | | Random Write | 0.080 ms | 96.741 ms | 0.00000 | 0.00000 | | Read | 0.001 ms | 0.191 ms | 0.00000 | 0.00000 | | Random Read | 0.001 ms | 0.181 ms | 0.00000 | 0.00000 | |--------------+-----------------+-----------------+----------+-----------| | Total | 0.145 ms | 2558.833 ms | 0.00034 | 0.00000 | `--------------+-----------------+-----------------+----------+-----------'
