BHyVeHackathonに参加してきた
2012年7月28日，29日にBHyVe Hackathonというやつがあったので
参加してみました．
と言っても大したことできないのでとりあえず，bhyve上のゲスト
のベンチマークとってみました．ゲストとホストのdmesgをここに
でベンチマーク結果です．
ベンチマークとして3つ，Bonnie++，iperf，UnixBenchを使って
測ってみました．
以下に示します．
まずゲストの方です．
=====BONNIE=====
bhyve-tap0% bonnie++ -s 512 -r 256
Writing a byte at a time...done
Writing intelligently...done
Rewriting...done
Reading a byte at a time...done
Reading intelligently...done
start 'em...done...done...done...done...done...
Create files in sequential order...done.
Stat files in sequential order...done.
Delete files in sequential order...done.
Create files in random order...done.
Stat files in random order...done.
Delete files in random order...done.
Version 1.96 ------Sequential Output------ --Sequential Input- --Random-
Concurrency 1 -Per Chr- --Block-- -Rewrite- -Per Chr- --Block-- --Seeks--
Machine Size K/sec %CP K/sec %CP K/sec %CP K/sec %CP K/sec %CP /sec %CP
bhyve-tap0 512M 211 99 51131 9 47915 10 413 99 +++++ +++ +++++ +++
Latency 49818us 57872us 2123ms 27809us 407us 182ms
Version 1.96 ------Sequential Create------ --------Random Create--------
bhyve-tap0 -Create-- --Read--- -Delete-- -Create-- --Read--- -Delete--
files /sec %CP /sec %CP /sec %CP /sec %CP /sec %CP /sec %CP
16 4800 30 +++++ +++ 8281 39 4848 30 +++++ +++ 8163 39
Latency 11633us 112us 241us 5528us 91us 675us
1.96,1.96,bhyve-tap0,1,1336390335,512M,,211,99,51131,9,47915,10,413,99,+++++,+++,+++++,+++,16,,,,,4800,30,+++++,+++,8281,39,4848,30,+++++,+++,8163,39,49818us,57872us,2123ms,27809us,407us,182ms,11633us,112us,241us,5528us,91us,675us
=====IPERF CLINET=====
bhyve-tap0% iperf -c localhost
------------------------------------------------------------
Client connecting to localhost, TCP port 5001
TCP window size: 42.0 KByte (default)
------------------------------------------------------------
[ 3] local 127.0.0.1 port 31922 connected with 127.0.0.1 port 5001
[ ID] Interval Transfer Bandwidth
[ 3] 0.0-10.0 sec 3.60 GBytes 3.09 Gbits/sec
bhyve-tap0% iperf -c 192.168.168.170
------------------------------------------------------------
Client connecting to 192.168.168.170, TCP port 5001
TCP window size: 48.5 KByte (default)
------------------------------------------------------------
[ 3] local 192.168.168.176 port 46717 connected with 192.168.168.170 port 5001
[ ID] Interval Transfer Bandwidth
[ 3] 0.0-10.0 sec 459 MBytes 385 Mbits/sec
=====IPERF SERVER=====
bhyve-tap0% iperf -s
------------------------------------------------------------
Server listening on TCP port 5001
TCP window size: 64.0 KByte (default)
------------------------------------------------------------
[ 4] local 127.0.0.1 port 5001 connected with 127.0.0.1 port 31922
[ ID] Interval Transfer Bandwidth
[ 4] 0.0-10.0 sec 3.60 GBytes 3.08 Gbits/sec
[ 5] local 192.168.168.176 port 5001 connected with 192.168.168.170 port 49228
[ 5] 0.0-10.0 sec 476 MBytes 399 Mbits/sec
[ 4] local 192.168.168.176 port 5001 connected with 192.168.168.66 port 62849
[ 4] 0.0-10.0 sec 808 MBytes 677 Mbits/sec
=======UNIX BENCH=======
bhyve-tap0% unixbench
# # # # # # # ##### ###### # # #### # #
# # ## # # # # # # # ## # # # # #
# # # # # # ## ##### ##### # # # # ######
# # # # # # ## # # # # # # # # #
# # # ## # # # # # # # ## # # # #
#### # # # # # ##### ###### # # #### # #
4 1 Based on the Byte Magazine Unix Benchmark
44 11
v v 4 4 1
v v 44444 1 v4.1 revisions mostly by David C. Niemi,
v 4 o 111 Reston, VA, USA <niemi@tux.org>
Dhrystone 2 using register variables 1
2 3 4 5 6 7 8 9 10
Double-Precision Whetstone 1 2 3 4 5 6 7 8 9 10
System Call Overhead 1 2 3 4 5 6 7 8 9 10
Pipe Throughput 1 2 3 4 5 6 7 8 9 10
Pipe-based Context Switching 1 2 3 4 5 6 7 8 9 10
Process Creation 1 2 3
Execl Throughput 1 2 3
Filesystem Throughput 1024 bufsize 2000 maxblocks 1 2 3
Filesystem Throughput 256 bufsize 500 maxblocks 1 2 3
Filesystem Throughput 4096 bufsize 8000 maxblocks 1 2 3
Shell Scripts (1 concurrent) 1 2 3
Shell Scripts (8 concurrent) 1 2 3
Shell Scripts (16 concurrent) 1 2 3
Arithmetic Test (type = short) 1 2 3
Arithmetic Test (type = int) 1 2 3
Arithmetic Test (type = long) 1 2 3
Arithmetic Test (type = float) 1 2 3
Arithmetic Test (type = double) 1 2 3
Arithoh 1 2 3
C Compiler Throughput 1 2 3
Dc: sqrt(2) to 99 decimal places 1 2 3
Recursion Test--Tower of Hanoi 1 2 3
==============================================================
BYTE UNIX Benchmarks (Version 4.1.0)
System -- bhyve-tap0
Start Benchmark Run: Mon May 7 13:26:14 UTC 2012
3 interactive users.
1:26PM up 16:27, 3 users, load averages: 0.07, 0.02, 0.00
-r-xr-xr-x 1 root wheel 142880 Jan 3 07:55 /bin/sh
/bin/sh: ELF 64-bit LSB executable, x86-64, version 1 (FreeBSD), dynamically linked (uses shared libs), for FreeBSD 9.0 (900044), stripped
/dev/vtbd0 1007516 325292 601624 35% /
Dhrystone 2 using register variables 13402152.3 lps (10.0 secs, 10 samples)
Double-Precision Whetstone 2845.6 MWIPS (10.0 secs, 10 samples)
System Call Overhead 600086.8 lps (10.0 secs, 10 samples)
Pipe Throughput 763106.7 lps (10.0 secs, 10 samples)
Pipe-based Context Switching 63390.4 lps (10.0 secs, 10 samples)
Process Creation 3264.5 lps (30.0 secs, 3 samples)
Execl Throughput 1143.7 lps (30.0 secs, 3 samples)
File Read 1024 bufsize 2000 maxblocks 416090.0 KBps (30.0 secs, 3 samples)
File Write 1024 bufsize 2000 maxblocks 66560.0 KBps (30.0 secs, 3 samples)
File Copy 1024 bufsize 2000 maxblocks 90105.0 KBps (30.0 secs, 3 samples)
File Read 256 bufsize 500 maxblocks 106710.0 KBps (30.0 secs, 3 samples)
File Write 256 bufsize 500 maxblocks 65655.0 KBps (30.0 secs, 3 samples)
File Copy 256 bufsize 500 maxblocks 36586.0 KBps (30.0 secs, 3 samples)
File Read 4096 bufsize 8000 maxblocks 1165741.0 KBps (30.0 secs, 3 samples)
File Write 4096 bufsize 8000 maxblocks 39644.0 KBps (30.0 secs, 3 samples)
File Copy 4096 bufsize 8000 maxblocks 49622.0 KBps (30.0 secs, 3 samples)
Shell Scripts (1 concurrent) 2655.3 lpm (60.0 secs, 3 samples)
Shell Scripts (8 concurrent) 356.0 lpm (60.0 secs, 3 samples)
Shell Scripts (16 concurrent) 178.0 lpm (60.0 secs, 3 samples)
Arithmetic Test (type = short) 1928596.5 lps (10.0 secs, 3 samples)
Arithmetic Test (type = int) 1928929.0 lps (10.0 secs, 3 samples)
Arithmetic Test (type = long) 711373.1 lps (10.0 secs, 3 samples)
Arithmetic Test (type = float) 1498945.9 lps (10.0 secs, 3 samples)
Arithmetic Test (type = double) 962181.3 lps (10.0 secs, 3 samples)
Arithoh 303218242.9 lps (10.0 secs, 3 samples)
C Compiler Throughput 1034.7 lpm (60.0 secs, 3 samples)
Dc: sqrt(2) to 99 decimal places 103272.0 lpm (30.0 secs, 3 samples)
Recursion Test--Tower of Hanoi 156188.9 lps (20.0 secs, 3 samples)
INDEX VALUES
TEST BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 13402152.3 1148.4
Double-Precision Whetstone 55.0 2845.6 517.4
Execl Throughput 43.0 1143.7 266.0
File Copy 1024 bufsize 2000 maxblocks 3960.0 90105.0 227.5
File Copy 256 bufsize 500 maxblocks 1655.0 36586.0 221.1
File Copy 4096 bufsize 8000 maxblocks 5800.0 49622.0 85.6
Pipe Throughput 12440.0 763106.7 613.4
Pipe-based Context Switching 4000.0 63390.4 158.5
Process Creation 126.0 3264.5 259.1
Shell Scripts (8 concurrent) 6.0 356.0 593.3
System Call Overhead 15000.0 600086.8 400.1
=========
FINAL SCORE 323.5
bhyve-tap0%
次にhostのベンチマーク結果です．
=====BONNIE++======
-[33]% bonnie++ -m 512 -r 256
Writing a byte at a time...done
Writing intelligently...done
Rewriting...done
Reading a byte at a time...done
Reading intelligently...done
start 'em...done...done...done...done...done...
Create files in sequential order...done.
Stat files in sequential order...done.
Delete files in sequential order...done.
Create files in random order...done.
Stat files in random order...done.
Delete files in random order...done.
Version 1.96 ------Sequential Output------ --Sequential Input- --Random-
Concurrency 1 -Per Chr- --Block-- -Rewrite- -Per Chr- --Block-- --Seeks--
Machine Size K/sec %CP K/sec %CP K/sec %CP K/sec %CP K/sec %CP /sec %CP
512 512M 259 99 50808 9 50101 15 487 99 896333 99 2613 76
Latency 32495us 59089us 62541us 28319us 1240us 6932us
Version 1.96 ------Sequential Create------ --------Random Create--------
512 -Create-- --Read--- -Delete-- -Create-- --Read--- -Delete--
files /sec %CP /sec %CP /sec %CP /sec %CP /sec %CP /sec %CP
16 29350 74 +++++ +++ +++++ +++ 27305 84 +++++ +++ +++++ +++
Latency 138ms 46us 43us 84306us 45us 36us
1.96,1.96,512,1,1343489501,512M,,259,99,50808,9,50101,15,487,99,896333,99,2613,76,16,,,,,29350,74,+++++,+++,+++++,+++,27305,84,+++++,+++,+++++,+++,32495us,59089us,62541us,28319us,1240us,6932us,138ms,46us,43us,84306us,45us,36us
=====IPERF CLIENT======
$ iperf -c localhost
------------------------------------------------------------
Client connecting to localhost, TCP port 5001
TCP window size: 42.0 KByte (default)
------------------------------------------------------------
[ 3] local 127.0.0.1 port 50376 connected with 127.0.0.1 port 5001
[ ID] Interval Transfer Bandwidth
[ 3] 0.0-10.0 sec 8.22 GBytes 7.06 Gbits/sec
iperf -c localhost 0.02s user 9.87s system 98% cpu 10.073 total
-[36]% iperf -c 192.168.168.176
------------------------------------------------------------
Client connecting to 192.168.168.176, TCP port 5001
TCP window size: 32.5 KByte (default)
------------------------------------------------------------
[ 3] local 192.168.168.170 port 49228 connected with 192.168.168.176 port 5001
[ ID] Interval Transfer Bandwidth
[ 3] 0.0-10.0 sec 476 MBytes 399 Mbits/sec
-[37]% iperf -c 192.168.168.66
------------------------------------------------------------
Client connecting to 192.168.168.66, TCP port 5001
TCP window size: 32.5 KByte (default)
------------------------------------------------------------
[ 3] local 192.168.168.170 port 62802 connected with 192.168.168.66 port 5001
[ ID] Interval Transfer Bandwidth
[ 3] 0.0-10.0 sec 1020 MBytes 856 Mbits/sec
=====IPERF SERVER=====
-[34]% iperf -s
-[34]% iperf -s
------------------------------------------------------------
Server listening on TCP port 5001
TCP window size: 64.0 KByte (default)
------------------------------------------------------------
[ 4] local 127.0.0.1 port 5001 connected with 127.0.0.1 port 50376
[ ID] Interval Transfer Bandwidth
[ 4] 0.0-10.0 sec 8.22 GBytes 7.06 Gbits/sec
[ 5] local 192.168.168.170 port 5001 connected with 192.168.168.176 port 46717
[ 5] 0.0-10.0 sec 459 MBytes 384 Mbits/sec
[ 4] local 192.168.168.170 port 5001 connected with 192.168.168.66 port 62834
[ 4] 0.0-10.0 sec 1.05 GBytes 904 Mbits/sec
-[42]% unixbench
# # # # # # # ##### ###### # # #### # #
# # ## # # # # # # # ## # # # # #
# # # # # # ## ##### ##### # # # # ######
# # # # # # ## # # # # # # # # #
# # # ## # # # # # # # ## # # # #
#### # # # # # ##### ###### # # #### # #
4 1 Based on the Byte Magazine Unix Benchmark
44 11
v v 4 4 1
v v 44444 1 v4.1 revisions mostly by David C. Niemi,
v 4 o 111 Reston, VA, USA <niemi@tux.org>
Dhrystone 2 using register variables 1 2 3 4 5 6 7 8 9 10
Double-Precision Whetstone 1 2 3 4 5 6 7 8 9 10
System Call Overhead 1 2 3 4 5 6 7 8 9 10
Pipe Throughput 1 2 3 4 5 6 7 8 9 10
Pipe-based Context Switching 1 2 3 4 5 6 7 8 9 10
Process Creation 1 2 3
Execl Throughput 1 2 3
Filesystem Throughput 1024 bufsize 2000 maxblocks 1 2 3
Filesystem Throughput 256 bufsize 500 maxblocks 1 2 3
Filesystem Throughput 4096 bufsize 8000 maxblocks 1 2 3
Shell Scripts (1 concurrent) 1 2 3
Shell Scripts (8 concurrent) 1 2 3
Shell Scripts (16 concurrent) 1 2 3
Arithmetic Test (type = short) 1 2 3
Arithmetic Test (type = int) 1 2 3
Arithmetic Test (type = long) 1 2 3
Arithmetic Test (type = float) 1 2 3
Arithmetic Test (type = double) 1 2 3
Arithoh 1 2 3
C Compiler Throughput 1 2 3
Dc: sqrt(2) to 99 decimal places 1 2 3
Recursion Test--Tower of Hanoi 1 2 3
==============================================================
BYTE UNIX Benchmarks (Version 4.1.0)
System -- bee
Start Benchmark Run: Sun Jul 29 03:55:50 JST 2012
3 interactive users.
3:55AM up 17:57, 3 users, load averages: 1.00, 0.99, 0.92
-r-xr-xr-x 1 root wheel 143200 Jul 28 16:48 /bin/sh
/bin/sh: ELF 64-bit LSB executable, x86-64, version 1 (FreeBSD), dynamically linked (uses shared libs), FreeBSD-style, for FreeBSD 10.0 (1000014), stripped
/dev/ada0p2 149650652 7552928 130125672 5% /
Dhrystone 2 using register variables 14366223.6 lps (10.0 secs, 10 samples)
Double-Precision Whetstone 2639.9 MWIPS (9.9 secs, 10 samples)
System Call Overhead 713536.2 lps (10.0 secs, 10 samples)
Pipe Throughput 926352.8 lps (10.0 secs, 10 samples)
Pipe-based Context Switching 148104.0 lps (10.0 secs, 10 samples)
Process Creation 4459.8 lps (30.0 secs, 3 samples)
Execl Throughput 1782.5 lps (29.7 secs, 3 samples)
File Read 1024 bufsize 2000 maxblocks 486558.0 KBps (30.0 secs, 3 samples)
File Write 1024 bufsize 2000 maxblocks 84811.0 KBps (30.0 secs, 3 samples)
File Copy 1024 bufsize 2000 maxblocks 86390.0 KBps (30.0 secs, 3 samples)
File Read 256 bufsize 500 maxblocks 125491.0 KBps (30.0 secs, 3 samples)
File Write 256 bufsize 500 maxblocks 81021.0 KBps (30.0 secs, 3 samples)
File Copy 256 bufsize 500 maxblocks 44906.0 KBps (30.0 secs, 3 samples)
File Read 4096 bufsize 8000 maxblocks 1425878.0 KBps (30.0 secs, 3 samples)
File Write 4096 bufsize 8000 maxblocks 49866.0 KBps (30.0 secs, 3 samples)
File Copy 4096 bufsize 8000 maxblocks 44958.0 KBps (30.0 secs, 3 samples)
Shell Scripts (1 concurrent) 717.7 lpm (60.0 secs, 3 samples)
Shell Scripts (8 concurrent) 167.3 lpm (60.0 secs, 3 samples)
Shell Scripts (16 concurrent) 88.0 lpm (60.0 secs, 3 samples)
Arithmetic Test (type = short) 1936049.4 lps (10.0 secs, 3 samples)
Arithmetic Test (type = int) 1931551.3 lps (10.0 secs, 3 samples)
Arithmetic Test (type = long) 711919.7 lps (10.0 secs, 3 samples)
Arithmetic Test (type = float) 1521321.8 lps (10.0 secs, 3 samples)
Arithmetic Test (type = double) 972646.2 lps (10.0 secs, 3 samples)
Arithoh 302778793.2 lps (10.0 secs, 3 samples)
C Compiler Throughput 668.3 lpm (60.0 secs, 3 samples)
Dc: sqrt(2) to 99 decimal places 148313.2 lpm (30.0 secs, 3 samples)
Recursion Test--Tower of Hanoi 155617.2 lps (20.0 secs, 3 samples)
INDEX VALUES
TEST BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 14366223.6 1231.0
Double-Precision Whetstone 55.0 2639.9 480.0
Execl Throughput 43.0 1782.5 414.5
File Copy 1024 bufsize 2000 maxblocks 3960.0 86390.0 218.2
File Copy 256 bufsize 500 maxblocks 1655.0 44906.0 271.3
File Copy 4096 bufsize 8000 maxblocks 5800.0 44958.0 77.5
Pipe Throughput 12440.0 926352.8 744.7
Pipe-based Context Switching 4000.0 148104.0 370.3
Process Creation 126.0 4459.8 354.0
Shell Scripts (8 concurrent) 6.0 167.3 278.8
System Call Overhead 15000.0 713536.2 475.7
=========
FINAL SCORE 363.3とこんなかんじになっています．
CPUが思ったより下がってないけど，ネットワーク，ディスクは…
てな感じの結果でした．
あとはapacheのベンチマークとか取ればよかったかも
