Several things. One, you left out a ton of information, as this could be the result of many, many things. Where to start?
What kind of disk/controller for each machine/disk? What is the results you get from running bonnie? Can you give the output of hdparm for each disk? Also notice the the x86 box you are talking about is a considerably faster processor/memory combination, and no, I'm not talking about bogomips. Try the solaris test again, Binkey, and this time use a filesystem other than /tmp for the output. Try setting up a ufs-log file system and see if the performance doesn't exceed that of linux. That's your homework for today. Tomorrow we'll get to advanced disk performance considerations. ~:^) a "Tille, Andreas" wrote: > > Hello, > > I´ve got a Sparc E250 server and I´m runing Debian GNU/Linux on it. > I now started to make some performance comparisons against a default PC. > Here are the configurations I used: > > E250: > ~> cat /proc/cpuinfo > cpu : TI UltraSparc II (BlackBird) > fpu : UltraSparc II integrated FPU > promlib : Version 3 Revision 16 > prom : 3.16.1 > type : sun4u > ncpus probed : 2 > ncpus active : 2 > Cpu0Bogo : 799.53 > Cpu1Bogo : 799.53 > MMU Type : Spitfire > State: > CPU0: online > CPU1: online > > 2 GB RAM > 4x36 GB SCSI > > ~> uname -a > Linux bse 2.2.18pre21 #1 SMP Tue Dec 19 22:53:19 CET 2000 sparc64 unknown > > PC: > ~> cat /proc/cpuinfo > processor : 0 > vendor_id : GenuineIntel > cpu family : 6 > model : 8 > model name : Pentium III (Coppermine) > stepping : 6 > cpu MHz : 795.576 > cache size : 256 KB > fpu : yes > fpu_exception : yes > cpuid level : 2 > wp : yes > flags : fpu vme de pse tsc msr pae mce cx8 sep mtrr pge mca cmov > pat pse36 mmx fxsr sse > bogomips : 1589.24 > > 256 MB RAM > 20 GB IDE > > With this setup I noticed that the PC is about two times faster in > inserting data into a postgresql database. I´ve got a hint to test > disk speed and I did so with the following script: > > #!/bin/bash > date > ls -l /home/tillea/hd-test.file > i=0 > while [ $i -lt 100 ] ; do > echo -n "$i " > let i="$i+1" > dd if=/home/tillea/hd-test.file of=/tmp/hd-test.file 2> /dev/null > done > ls -l /tmp/hd-test.file > date > ### end of scrip > > Here are the results: > > PC: > > $ uname -a > Linux wr-linux02 2.4.5 #1 Mit Mai 30 08:24:06 CEST 2001 i686 unknown > > $ hd-test > Don Aug 23 12:21:56 CEST 2001 > -rw-r--r-- 1 tillea admin 25600000 23. Aug 11:50 > /home/tillea/hd-test.file > 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 > 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 > 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 > 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 -rw-r--r-- 1 > tillea admin 25600000 23. Aug 12:22 /tmp/hd-test.file > Don Aug 23 12:22:47 CEST 2001 > > ----> 51s for the PC > > Sparc: > > $ uname -a > SunOS bse 5.8 Generic_108528-08 sun4u sparc SUNW,Ultra-250 > > $ hd-test > Thu Aug 23 13:23:38 MEST 2001 > -rw-r--r-- 1 tillea admin 25600000 Aug 23 13:13 > /home/tillea/hd-test.file > 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 > 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 > 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 > 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 -rw-r--r-- 1 > tillea admin 25600000 Aug 23 13:26 /tmp/hd-test.file > Thu Aug 23 13:26:49 MEST 2001 > > -----> 191s for the Sparc running Solaris 8 > > $ uname -a > Linux bse 2.2.18pre21 #1 SMP Tue Dec 19 22:53:19 CET 2000 sparc64 unknown > > Don Aug 23 13:55:37 CEST 2001 > -rw-r--r-- 1 tillea admin 25600000 23. Aug 11:50 > /home/tillea/hd-test.file > 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 > 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 > 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 > 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 -rw-r--r-- 1 > tillea admin 25600000 23. Aug 13:57 /tmp/hd-test.file > Don Aug 23 13:57:16 CEST 2001 > > -----> 99s for tge Sparc running Debian GNU/Linux > > So the relation is about: > > PC/linux : Sparc/Solaris : Sparc/Linux > > 1 : 3.7 : 1.9 > > That´s not good for the Sparc machine (and even worse for Solaris by the way > but this is of less interest because the machine should run under Linux). > > Any hint if there some optimisations could be done?
