UnixBench results

[matta]

Here are the UnixBench 4.1.0 results for a 128MB VM. This is while the host is mainly idle, 2 reloads (rsync) were running which does use a lot of sequential disk I/O.

BYTE UNIX Benchmarks (Version 4.1.0)
System -- Linux xxx.com 2.6.10-xenU #3 Sun Jan 30 14:21:15 EST 2005 i686 AMD Opteron(tm) Processor 240 unknown GNU/Linux
Start Benchmark Run: Sat Feb 12 19:37:10 CET 2005
1 interactive users.
19:37:10 up 1 day, 9:18, 1 user, load average: 0.08, 0.03, 0.01
lrwxrwxrwx 1 root root 4 Feb 5 08:15 /bin/sh -> bash
/bin/sh: symbolic link to `bash'
/dev/sda1 6047492 271868 5468424 5% /
Dhrystone 2 using register variables 3063414.2 lps (10.0 secs, 1 samples)
Double-Precision Whetstone 735.6 MWIPS (10.8 secs, 1 samples)
Execl Throughput 1488.5 lps (29.2 secs, 1 samples)
File Read 1024 bufsize 2000 maxblocks 653647.0 KBps (30.0 secs, 1 samples)
File Write 1024 bufsize 2000 maxblocks 320400.0 KBps (30.0 secs, 1 samples)
File Copy 1024 bufsize 2000 maxblocks 236746.0 KBps (30.0 secs, 1 samples)
File Read 256 bufsize 500 maxblocks 389862.0 KBps (30.0 secs, 1 samples)
File Write 256 bufsize 500 maxblocks 174016.0 KBps (30.0 secs, 1 samples)
File Copy 256 bufsize 500 maxblocks 110320.0 KBps (30.0 secs, 1 samples)
File Read 4096 bufsize 8000 maxblocks 1008479.0 KBps (30.0 secs, 1 samples)
File Write 4096 bufsize 8000 maxblocks 615200.0 KBps (30.0 secs, 1 samples)
File Copy 4096 bufsize 8000 maxblocks 345582.0 KBps (30.0 secs, 1 samples)
Pipe Throughput 797744.2 lps (10.0 secs, 1 samples)
Pipe-based Context Switching 127448.1 lps (10.0 secs, 1 samples)
Process Creation 3710.1 lps (30.0 secs, 1 samples)
System Call Overhead 755149.3 lps (10.0 secs, 1 samples)
Shell Scripts (8 concurrent) 271.8 lpm (60.0 secs, 1 samples)


INDEX VALUES
TEST BASELINE RESULT INDEX

Dhrystone 2 using register variables 116700.0 3063414.2 262.5
Double-Precision Whetstone 55.0 735.6 133.7
Execl Throughput 43.0 1488.5 346.2
File Copy 1024 bufsize 2000 maxblocks 3960.0 236746.0 597.8
File Copy 256 bufsize 500 maxblocks 1655.0 110320.0 666.6
File Copy 4096 bufsize 8000 maxblocks 5800.0 345582.0 595.8
Pipe Throughput 12440.0 797744.2 641.3
Process Creation 126.0 3710.1 294.5
Shell Scripts (8 concurrent) 6.0 271.8 453.0
System Call Overhead 15000.0 755149.3 503.4
=========
FINAL SCORE 406.9

[GameraFan]

My results for a 192MB plan:

11:24:29 up 19:26, 1 user, load average: 0.33, 0.10, 0.02
lrwxrwxrwx 1 root root 4 Feb 15 15:48 /bin/sh -> bash
/bin/sh: symbolic link to bash
/dev/sda1 11308976 913364 9821148 9% /
Dhrystone 2 using register variables 3379754.9 lps (10.0 secs, 10 samples)
Double-Precision Whetstone 717.5 MWIPS (10.0 secs, 10 samples)
System Call Overhead 719619.6 lps (10.0 secs, 10 samples)
Pipe Throughput 778838.9 lps (10.0 secs, 10 samples)
Pipe-based Context Switching 113103.2 lps (10.0 secs, 10 samples)
Process Creation 3311.4 lps (30.0 secs, 3 samples)
Execl Throughput 2017.4 lps (29.7 secs, 3 samples)
File Read 1024 bufsize 2000 maxblocks 536738.0 KBps (30.0 secs, 3 samples)
File Write 1024 bufsize 2000 maxblocks 342066.0 KBps (30.0 secs, 3 samples)
File Copy 1024 bufsize 2000 maxblocks 206764.0 KBps (30.0 secs, 3 samples)
File Read 256 bufsize 500 maxblocks 335953.0 KBps (30.0 secs, 3 samples)
File Write 256 bufsize 500 maxblocks 154983.0 KBps (30.0 secs, 3 samples)
File Copy 256 bufsize 500 maxblocks 95690.0 KBps (30.0 secs, 3 samples)
File Read 4096 bufsize 8000 maxblocks 694538.0 KBps (30.0 secs, 3 samples)
File Write 4096 bufsize 8000 maxblocks 516570.0 KBps (30.0 secs, 3 samples)
File Copy 4096 bufsize 8000 maxblocks 280180.0 KBps (30.0 secs, 3 samples)
Shell Scripts (1 concurrent) 2855.3 lpm (60.0 secs, 3 samples)
Shell Scripts (8 concurrent) 386.8 lpm (60.0 secs, 3 samples)
Shell Scripts (16 concurrent) 196.3 lpm (60.0 secs, 3 samples)
Arithmetic Test (type = short) 263749.4 lps (10.0 secs, 3 samples)
Arithmetic Test (type = int) 265747.5 lps (10.0 secs, 3 samples)
Arithmetic Test (type = long) 264077.8 lps (10.0 secs, 3 samples)
Arithmetic Test (type = float) 492858.1 lps (10.0 secs, 3 samples)
Arithmetic Test (type = double) 499544.4 lps (10.0 secs, 3 samples)
Arithoh 5915302.7 lps (10.0 secs, 3 samples)
C Compiler Throughput 901.5 lpm (60.0 secs, 3 samples)
Dc: sqrt(2) to 99 decimal places 169542.5 lpm (30.0 secs, 3 samples)
Recursion Test--Tower of Hanoi 64011.7 lps (20.0 secs, 3 samples)


INDEX VALUES
TEST BASELINE RESULT INDEX

Dhrystone 2 using register variables 116700.0 3379754.9 289.6
Double-Precision Whetstone 55.0 717.5 130.5
Execl Throughput 43.0 2017.4 469.2
File Copy 1024 bufsize 2000 maxblocks 3960.0 206764.0 522.1
File Copy 256 bufsize 500 maxblocks 1655.0 95690.0 578.2
File Copy 4096 bufsize 8000 maxblocks 5800.0 280180.0 483.1
Pipe Throughput 12440.0 778838.9 626.1
Pipe-based Context Switching 4000.0 113103.2 282.8
Process Creation 126.0 3311.4 262.8
Shell Scripts (8 concurrent) 6.0 386.8 644.7
System Call Overhead 15000.0 719619.6 479.7
=========
FINAL SCORE 395.7

[matta]

By comparision a Dual Xeon 3.06Ghz w/ 1GB+ RAM scores around a 350.

[matta]

The non-beta servers are Opteron 242... even better result in a 128MB VM.

INDEX VALUES
TEST BASELINE RESULT INDEX

Dhrystone 2 using register variables 116700.0 4248277.0 364.0
Double-Precision Whetstone 55.0 1075.8 195.6
Execl Throughput 43.0 2269.7 527.8
File Copy 1024 bufsize 2000 maxblocks 3960.0 290736.0 734.2
File Copy 256 bufsize 500 maxblocks 1655.0 115869.0 700.1
File Copy 4096 bufsize 8000 maxblocks 5800.0 487190.0 840.0
Pipe Throughput 12440.0 1027625.6 826.1
Process Creation 126.0 3998.4 317.3
Shell Scripts (8 concurrent) 6.0 254.0 423.3
System Call Overhead 15000.0 1039846.3 693.2
=========
FINAL SCORE 512.4

[evanchsa]

Hah!

I think that it is very cool that you guys are benchmarking Xen performance. You even link to the paper that I helped write!

http://www.clarkson.edu/class/cs644/xen/files/repeatedxen-usenix04.pdf


Very cool!