Which Device shall i buy? (Pogoplug/GoFlex/Raspberry)

[tedm]

Thanks for posting that moonman, I'm running it now.

rty, I think it is already compiled, I ran make and make Makefile, and it was all up to date.

Ran ./Run from the ~/root/UnixBench dir and got the logo, and went right into the Dyrystone test (where it is now)
there is one .sh: 3dinfo: command not found error line above the ascii "UNIX Benchmarks" logo, but I didn't disable graphics.

I also have 60 processes running, including nginx, fastphp, samba, and mysql, so my results may not be optimal...

[tedm]

here are my dockstar results:

$this->bbcode_second_pass_code('', '========================================================================
BYTE UNIX Benchmarks (Version 5.1.3)

System: tedm_linux: GNU/Linux
OS: GNU/Linux -- 3.1.10-16-ARCH -- #1 PREEMPT Tue Feb 12 01:17:10 UTC 2013
Machine: armv5tel (unknown)
Language: en_US.utf8 (charmap="UTF-8", collate="UTF-8")
17:41:26 up 21:38, 1 user, load average: 0.81, 0.81, 0.53; runlevel 5

------------------------------------------------------------------------
Benchmark Run: Sun Mar 03 2013 17:41:27 - 18:18:55
0 CPUs in system; running 1 parallel copy of tests

Dhrystone 2 using register variables 2353143.2 lps (10.0 s, 7 samples)
Double-Precision Whetstone 46.4 MWIPS (9.9 s, 7 samples)
Execl Throughput 536.3 lps (29.2 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 74374.5 KBps (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 24284.5 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 5681.4 KBps (32.4 s, 2 samples)
Pipe Throughput 206851.9 lps (10.2 s, 7 samples)
Pipe-based Context Switching 17801.9 lps (10.0 s, 7 samples)
Process Creation 1337.5 lps (30.4 s, 2 samples)
Shell Scripts (1 concurrent) 367.9 lpm (60.5 s, 2 samples)
Shell Scripts (8 concurrent) 51.0 lpm (60.6 s, 2 samples)
System Call Overhead 599722.3 lps (10.1 s, 7 samples)

System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 2353143.2 201.6
Double-Precision Whetstone 55.0 46.4 8.4
Execl Throughput 43.0 536.3 124.7
File Copy 1024 bufsize 2000 maxblocks 3960.0 74374.5 187.8
File Copy 256 bufsize 500 maxblocks 1655.0 24284.5 146.7
File Copy 4096 bufsize 8000 maxblocks 5800.0 5681.4 9.8
Pipe Throughput 12440.0 206851.9 166.3
Pipe-based Context Switching 4000.0 17801.9 44.5
Process Creation 126.0 1337.5 106.2
Shell Scripts (1 concurrent) 42.4 367.9 86.8
Shell Scripts (8 concurrent) 6.0 51.0 85.0
System Call Overhead 15000.0 599722.3 399.8
========
System Benchmarks Index Score 84.5

[root@tedm_linux UnixBench]# ^C
')

[rty]

$this->bbcode_second_pass_quote('tedm', 'r')ty, I think it is already compiled, I ran make and make Makefile, and it was all up to date.

[tedm]

Here are the benchmarks for my dockstar with mysql, nginx, php-fpm, webmin, and samba shutdown:

$this->bbcode_second_pass_code('', '
========================================================================
BYTE UNIX Benchmarks (Version 5.1.3)

System: tedm_linux: GNU/Linux
OS: GNU/Linux -- 3.1.10-16-ARCH -- #1 PREEMPT Tue Feb 12 01:17:10 UTC 2013
Machine: armv5tel (unknown)
Language: en_US.utf8 (charmap="UTF-8", collate="UTF-8")
06:07:58 up 1 day, 10:04, 1 user, load average: 0.44, 0.43, 0.28; runlevel 5

------------------------------------------------------------------------
Benchmark Run: Mon Mar 04 2013 06:07:59 - 06:37:36
0 CPUs in system; running 1 parallel copy of tests

Dhrystone 2 using register variables 2398369.9 lps (10.0 s, 7 samples)
Double-Precision Whetstone 46.4 MWIPS (9.9 s, 7 samples)
Execl Throughput 584.1 lps (29.9 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 77762.0 KBps (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 25257.5 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 159379.5 KBps (30.0 s, 2 samples)
Pipe Throughput 215581.9 lps (10.2 s, 7 samples)
Pipe-based Context Switching 17858.0 lps (10.1 s, 7 samples)
Process Creation 1549.8 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 385.3 lpm (60.1 s, 2 samples)
Shell Scripts (8 concurrent) 54.5 lpm (60.6 s, 2 samples)
System Call Overhead 609276.9 lps (10.1 s, 7 samples)

System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 2398369.9 205.5
Double-Precision Whetstone 55.0 46.4 8.4
Execl Throughput 43.0 584.1 135.8
File Copy 1024 bufsize 2000 maxblocks 3960.0 77762.0 196.4
File Copy 256 bufsize 500 maxblocks 1655.0 25257.5 152.6
File Copy 4096 bufsize 8000 maxblocks 5800.0 159379.5 274.8
Pipe Throughput 12440.0 215581.9 173.3
Pipe-based Context Switching 4000.0 17858.0 44.6
Process Creation 126.0 1549.8 123.0
Shell Scripts (1 concurrent) 42.4 385.3 90.9
Shell Scripts (8 concurrent) 6.0 54.5 90.8
System Call Overhead 15000.0 609276.9 406.2
========
System Benchmarks Index Score 116.4

')

[rty]

Aha!!! From 84.5 to 116.4! It's quite significant.

Moonman, did you do the same (shutting down those daemons ) when you performed the test on yours?

[moonman]

I had samba running. It's pointless to compare performance of the same CPU. I was trying to make a point about Raspberry Pi vs Kirkwood 1.2GHz. It would be interesting to see results from other devices though.

[rty]

Result for Pogoplug series 4
$this->bbcode_second_pass_code('', '


sh: 3dinfo: command not found

# # # # # # # ##### ###### # # #### # #
# # ## # # # # # # # ## # # # # #
# # # # # # ## ##### ##### # # # # ######
# # # # # # ## # # # # # # # # #
# # # ## # # # # # # # ## # # # #
#### # # # # # ##### ###### # # #### # #

Version 5.1.3 Based on the Byte Magazine Unix Benchmark

Multi-CPU version Version 5 revisions by Ian Smith,
Sunnyvale, CA, USA
January 13, 2011 johantheghost at yahoo period com


1 x Dhrystone 2 using register variables 1
1 2 3 4 5 6 7 8 9 10

1 x Double-Precision Whetstone 1 2 3 4 5 6 7 8 9 10

1 x Execl Throughput 1 2 3

1 x File Copy 1024 bufsize 2000 maxblocks 1 2 3

1 x File Copy 256 bufsize 500 maxblocks 1 2 3

1 x File Copy 4096 bufsize 8000 maxblocks 1 2 3

1 x Pipe Throughput 1 2 3 4 5 6 7 8 9 10

1 x Pipe-based Context Switching 1 2 3 4 5 6 7 8 9 10

1 x Process Creation 1 2 3

1 x System Call Overhead 1 2 3 4 5 6 7 8 9 10

1 x Shell Scripts (1 concurrent) 1 2 3

1 x Shell Scripts (8 concurrent) 1 2 3

========================================================================
BYTE UNIX Benchmarks (Version 5.1.3)

System: pogo4: GNU/Linux
OS: GNU/Linux -- 3.7.8-0-ARCH -- #1 PREEMPT Sat Feb 16 01:51:44 UTC 2013
Machine: armv5tel (unknown)
Language: en_US.utf8 (charmap="UTF-8", collate="UTF-8")
16:59:11 up 9 days, 7:57, 1 user, load average: 0.04, 0.04, 0.05; runlevel 5

------------------------------------------------------------------------
Benchmark Run: Mon Mar 04 2013 16:59:11 - 17:27:44
0 CPUs in system; running 1 parallel copy of tests

Dhrystone 2 using register variables 1604503.7 lps (10.0 s, 7 samples)
Double-Precision Whetstone 31.2 MWIPS (9.9 s, 7 samples)
Execl Throughput 430.1 lps (29.5 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 47933.0 KBps (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 14666.4 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 108612.9 KBps (30.0 s, 2 samples)
Pipe Throughput 135430.2 lps (10.0 s, 7 samples)
Pipe-based Context Switching 12999.9 lps (10.0 s, 7 samples)
Process Creation 1179.0 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 273.7 lpm (60.1 s, 2 samples)
Shell Scripts (8 concurrent) 37.8 lpm (60.3 s, 2 samples)
System Call Overhead 404167.4 lps (10.0 s, 7 samples)

System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 1604503.7 137.5
Double-Precision Whetstone 55.0 31.2 5.7
Execl Throughput 43.0 430.1 100.0
File Copy 1024 bufsize 2000 maxblocks 3960.0 47933.0 121.0
File Copy 256 bufsize 500 maxblocks 1655.0 14666.4 88.6
File Copy 4096 bufsize 8000 maxblocks 5800.0 108612.9 187.3
Pipe Throughput 12440.0 135430.2 108.9
Pipe-based Context Switching 4000.0 12999.9 32.5
Process Creation 126.0 1179.0 93.6
Shell Scripts (1 concurrent) 42.4 273.7 64.5
Shell Scripts (8 concurrent) 6.0 37.8 63.0
System Call Overhead 15000.0 404167.4 269.4
========
System Benchmarks Index Score 78.7
')

I wonder what the folks at Pogoplug were thinking when they designed its latest series 4...
Note: Samba copy speed is OK though. Copying a 1.5Gb file from desktop to P4 takes 23GB/s and from P4 to desktop 40GB/s. HDD mounted is the 2.5" 1TB Seagate on the SATA port.

I hope they show repentance and release the next Pogoplug 5 with the same specs as series 4 but faster CPU, at least 1.6Ghz, 1GB of RAM.

[tedm]

that's interesting rty. I think the next step in the arm ecosystem would be towards the tablet / chromeos family of multi-core SoCs. Looking forward to some benchmarks on say the samsung exynos running any linux, which can still run single digit watts with the display off, and fanless. We have the javascript browser performance data, but the byte tests run in a bash shell would be very interesting.

[pepedog]

Cubox with samba4 and minidlna running
$this->bbcode_second_pass_code('', 'sh: 3dinfo: command not found

# # # # # # # ##### ###### # # #### # #
# # ## # # # # # # # ## # # # # #
# # # # # # ## ##### ##### # # # # ######
# # # # # # ## # # # # # # # # #
# # # ## # # # # # # # ## # # # #
#### # # # # # ##### ###### # # #### # #

Version 5.1.3 Based on the Byte Magazine Unix Benchmark

Multi-CPU version Version 5 revisions by Ian Smith,
Sunnyvale, CA, USA
January 13, 2011 johantheghost at yahoo period com


1 x Dhrystone 2 using register variables 1 2 3 4 5 6 7 8 9 10

1 x Double-Precision Whetstone 1 2 3 4 5 6 7 8 9 10

1 x Execl Throughput 1 2 3

1 x File Copy 1024 bufsize 2000 maxblocks 1 2 3

1 x File Copy 256 bufsize 500 maxblocks 1 2 3

1 x File Copy 4096 bufsize 8000 maxblocks 1 2 3

1 x Pipe Throughput 1 2 3 4 5 6 7 8 9 10

1 x Pipe-based Context Switching 1 2 3 4 5 6 7 8 9 10

1 x Process Creation 1 2 3

1 x System Call Overhead 1 2 3 4 5 6 7 8 9 10

1 x Shell Scripts (1 concurrent) 1 2 3

1 x Shell Scripts (8 concurrent) 1 2 3

========================================================================
BYTE UNIX Benchmarks (Version 5.1.3)

System: alarmcubox: GNU/Linux
OS: GNU/Linux -- 3.5.7-9-ARCH+ -- #1 PREEMPT Fri Feb 15 02:16:10 UTC 2013
Machine: armv7l (unknown)
Language: en_US.utf8 (charmap="UTF-8", collate="ANSI_X3.4-1968")
20:26:59 up 14 min, 1 user, load average: 0.76, 0.43, 0.24; runlevel 5

------------------------------------------------------------------------
Benchmark Run: Mon Mar 04 2013 20:26:59 - 20:55:11
0 CPUs in system; running 1 parallel copy of tests

Dhrystone 2 using register variables 2564478.8 lps (10.0 s, 7 samples)
Double-Precision Whetstone 268.3 MWIPS (10.0 s, 7 samples)
Execl Throughput 687.5 lps (29.9 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 73115.3 KBps (30.1 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 24237.5 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 153653.6 KBps (30.0 s, 2 samples)
Pipe Throughput 183762.9 lps (10.0 s, 7 samples)
Pipe-based Context Switching 50057.4 lps (10.0 s, 7 samples)
Process Creation 2037.1 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 420.3 lpm (60.1 s, 2 samples)
Shell Scripts (8 concurrent) 59.7 lpm (60.3 s, 2 samples)
System Call Overhead 486824.1 lps (10.0 s, 7 samples)

System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 2564478.8 219.7
Double-Precision Whetstone 55.0 268.3 48.8
Execl Throughput 43.0 687.5 159.9
File Copy 1024 bufsize 2000 maxblocks 3960.0 73115.3 184.6
File Copy 256 bufsize 500 maxblocks 1655.0 24237.5 146.5
File Copy 4096 bufsize 8000 maxblocks 5800.0 153653.6 264.9
Pipe Throughput 12440.0 183762.9 147.7
Pipe-based Context Switching 4000.0 50057.4 125.1
Process Creation 126.0 2037.1 161.7
Shell Scripts (1 concurrent) 42.4 420.3 99.1
Shell Scripts (8 concurrent) 6.0 59.7 99.5
System Call Overhead 15000.0 486824.1 324.5
========
System Benchmarks Index Score 148.8')

[pepedog]

Trimslice
$this->bbcode_second_pass_code('', 'sh: 3dinfo: command not found

# # # # # # # ##### ###### # # #### # #
# # ## # # # # # # # ## # # # # #
# # # # # # ## ##### ##### # # # # ######
# # # # # # ## # # # # # # # # #
# # # ## # # # # # # # ## # # # #
#### # # # # # ##### ###### # # #### # #

Version 5.1.3 Based on the Byte Magazine Unix Benchmark

Multi-CPU version Version 5 revisions by Ian Smith,
Sunnyvale, CA, USA
January 13, 2011 johantheghost at yahoo period com


1 x Dhrystone 2 using register variables 1 2 3 4 5 6 7 8 9 10

1 x Double-Precision Whetstone 1 2 3 4 5 6 7 8 9 10

1 x Execl Throughput 1 2 3

1 x File Copy 1024 bufsize 2000 maxblocks 1 2 3

1 x File Copy 256 bufsize 500 maxblocks 1 2 3

1 x File Copy 4096 bufsize 8000 maxblocks 1 2 3

1 x Pipe Throughput 1 2 3 4 5 6 7 8 9 10

1 x Pipe-based Context Switching 1 2 3 4 5 6 7 8 9 10

1 x Process Creation 1 2 3

1 x System Call Overhead 1 2 3 4 5 6 7 8 9 10

1 x Shell Scripts (1 concurrent) 1 2 3

1 x Shell Scripts (8 concurrent) 1 2 3

========================================================================
BYTE UNIX Benchmarks (Version 5.1.3)

System: alarm: GNU/Linux
OS: GNU/Linux -- 3.1.10-5-ARCH -- #1 SMP PREEMPT Thu Nov 15 08:49:53 CST 2012
Machine: armv7l (unknown)
Language: en_US.utf8 (charmap="UTF-8", collate="UTF-8")
15:15:33 up 1 min, 1 user, load average: 0.51, 0.20, 0.07; runlevel 5

------------------------------------------------------------------------
Benchmark Run: Mon Mar 04 2013 15:15:33 - 15:44:09
0 CPUs in system; running 1 parallel copy of tests

Dhrystone 2 using register variables 3891766.5 lps (10.0 s, 7 samples)
Double-Precision Whetstone 581.7 MWIPS (10.0 s, 7 samples)
Execl Throughput 689.8 lps (29.9 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 75490.0 KBps (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 22741.0 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 192632.5 KBps (30.0 s, 2 samples)
Pipe Throughput 165967.5 lps (10.0 s, 7 samples)
Pipe-based Context Switching 36039.0 lps (10.0 s, 7 samples)
Process Creation 1936.1 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 636.2 lpm (60.0 s, 2 samples)
Shell Scripts (8 concurrent) 136.6 lpm (60.2 s, 2 samples)
System Call Overhead 455177.8 lps (10.0 s, 7 samples)

System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 3891766.5 333.5
Double-Precision Whetstone 55.0 581.7 105.8
Execl Throughput 43.0 689.8 160.4
File Copy 1024 bufsize 2000 maxblocks 3960.0 75490.0 190.6
File Copy 256 bufsize 500 maxblocks 1655.0 22741.0 137.4
File Copy 4096 bufsize 8000 maxblocks 5800.0 192632.5 332.1
Pipe Throughput 12440.0 165967.5 133.4
Pipe-based Context Switching 4000.0 36039.0 90.1
Process Creation 126.0 1936.1 153.7
Shell Scripts (1 concurrent) 42.4 636.2 150.0
Shell Scripts (8 concurrent) 6.0 136.6 227.7
System Call Overhead 15000.0 455177.8 303.5
========
System Benchmarks Index Score 177.0
')