Re: Fastest disk in the west or bad iozone numbers?

2006-08-03 Thread Dan Nelson 
In the last episode (Aug 03), Nikolas Britton said:
> Anyone have a clue why iozone reports disk read rates of 688
> MegaBytes/s on a 1GB test file? Am I doing something stupid, like not
> converting the numbers correctly?

You want to test using a file at least twice as big as your RAM,
otherwise you're just testing your cache.  You can see an example of
this on the first graph at http://www.iozone.org .  Only the far right
edge shows the disks speed.  web.archive.org shows that same image
existed back in 2000 with a modtime of 1999, so it's possible the
machine being tested had under 256MB of RAM.

-- 
Dan Nelson
[EMAIL PROTECTED]
___
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "[EMAIL PROTECTED]"

Fastest disk in the west or bad iozone numbers?

2006-08-02 Thread Nikolas Britton 

Anyone have a clue why iozone reports disk read rates of
688MegaBytes/s on a 1GB test file? Am I doing something stupid, like
not converting the numbers correctly?

I've attached the iozone report, It's from a single 400GB Seagate SATA drive.
infomatic# iozone -az -g 1g -b ~/seagate.xls
Iozone: Performance Test of File I/O
Version $Revision: 3.263 $
Compiled for 32 bit mode.
Build: freebsd

Contributors:William Norcott, Don Capps, Isom Crawford, Kirby Collins
 Al Slater, Scott Rhine, Mike Wisner, Ken Goss
 Steve Landherr, Brad Smith, Mark Kelly, Dr. Alain CYR,
 Randy Dunlap, Mark Montague, Dan Million,
 Jean-Marc Zucconi, Jeff Blomberg,
 Erik Habbinga, Kris Strecker, Walter Wong.

Run began: Wed Aug  2 13:22:54 2006

Auto Mode
Cross over of record size disabled.
Using maximum file size of 1048576 kilobytes.
Command line used: iozone -az -g 1g -b /root/seagate.xls
Output is in Kbytes/sec
Time Resolution = 0.02 seconds.
Processor cache size set to 1024 Kbytes.
Processor cache line size set to 32 bytes.
File stride size set to 17 * record size.

iozone test complete.
Excel output is below:

"Writer report"
"4"  "8"  "16"  "32"  "64"  "128"  "256"  "512"  "1024"  "2048"  "4096" 
 "8192"  "16384"
"64"   117858  208475  293496  467159  279442
"128"   13911  14740  32152  25182  67120  84273
"256"   27914  63615  235955  248516  27902  41006  71708
"512"   36384  64875  45791  275720  23293  50131  337516  82223
"1024"   28623  6571  97246  251659  71809  85375  125828  325509  278322
"2048"   47798  52503  41486  83218  79279  66456  47005  113475  74461  101551
"4096"   28189  12285  27437  51167  47354  53293  60537  50633  43507  44217  
59572
"8192"   30163  42406  66024  43097  26781  34145  33147  56543  32099  70854  
46321  66291
"16384"   28205  48073  39713  44404  48775  55348  67135  64065  61474  47064  
45921  45291  67852
"32768"   32041  58188  66444  69155  64704  69603  61746  64296  65845  57059  
54606  55612  64872
"65536"   35468  51383  58800  60465  61065  60876  60873  57916  57607  59288  
58676  56887  59718
"131072"   36040  53350  56741  56059  56369  57402  56329  55548  53857  54671 
 53841  58136  56973
"262144"   41989  54045  56191  56397  56645  56980  54622  56138  55409  57708 
 56486  55746  55302
"524288"   43288  53420  55245  55694  56406  56120  55809  55519  54507  55516 
 55370  55267  56061
"1048576"   46697  52919  54481  54439  55033  54735  54715  54120  55171  
54381  54624  54375  54315

"Re-writer report"
"4"  "8"  "16"  "32"  "64"  "128"  "256"  "512"  "1024"  "2048"  "4096" 
 "8192"  "16384"
"64"   653745  829499  943164  926881  633683
"128"   18127  67542  124981  202546  191015  469042
"256"   266396  325611  368399  133756  275911  349683  527997
"512"   23810  581247  111862  158560  14502  291555  581877  528987
"1024"   54540  62530  122578  385696  290659  370491  369535  466509  552932
"2048"   54979  64220  31940  70530  81616  35326  48416  79781  108400  71254
"4096"   50653  38137  55127  38932  82662  49606  86061  60064  47201  26341  
63558
"8192"   30646  27428  66534  81869  46981  42537  62341  60307  52390  36061  
46217  69476
"16384"   43660  54881  35682  48482  32834  41479  73412  61000  59474  62223  
56970  53378  71564
"32768"   53460  70274  70062  66694  63557  68102  7  57039  69162  61680  
64322  72151  52286
"65536"   53877  45904  56165  59962  59226  59401  57805  57885  57730  60035  
59937  59053  60268
"131072"   51140  56017  56050  56288  56041  55215  56123  55062  56726  56384 
 57301  55848  57552
"262144"   48664  55599  56843  57017  56667  57204  55478  57366  57193  56061 
 57085  56457  56565
"524288"   48505  55655  55984  55376  55646  56182  55536  56360  56130  56665 
 56682  56507  55632
"1048576"   50267  52582  53160  53615  53812  53592  53757  53218  53713  
53887  54246  54344  54261

"Reader report"
"4"  "8"  "16"  "32"  "64"  "128"  "256"  "512"  "1024"  "2048"  "4096" 
 "8192"  "16384"
"64"   943164  1144170  1365309  1256653  1280633
"128"   636395  1153045  1468419  1488779  1307491  1346852
"256"   885478  1163807  1361560  1407980  1375513  1422906  1319723
"512"   873453  1069138  1330031  1330031  1414110  1422540  1295529  980760
"1024"   920953  1143004  1245780  1426135  1426135  1190859  1326585  889667  
865465
"2048"   881234  1228430  1436297  1471229  1476539  1519371  1353467  837257  
799687  842512
"4096"   927107  1229979  1448358  1494862  1506925  1531235  1355348  820840  
822963  780774  824148
"8192"   879832  1244800  1471785  1506701  1528892  1553009  1348254  799535  
777007  758717  767907  778186
"16384"   895302  1142931  1481929  1518440  1530207  1416206  1242720  765569  
788939  781013  780454