UNIX bench better on 2.2 than 2.4?

UNIX bench better on 2.2 than 2.4?

Post by Jean-Eric Cuende » Sat, 16 Mar 2002 16:40:09



Hi,
I ran UNIX bench on 4 of our machines. The results are below.
What is strange is that 2.2 makes better than 2.4! Is it a known fact?
And is the UNIX bench better with 2 CPU?
-jec

Results:
Host1: 2xPIII 550MHz / 1Gb RAM / RAID5 SCSI
      Result: 164.7
Host2: 2xPIII 866MHz / 1Gb RAM / RAID1 soft IDE
      Result: 195.7
Host3: 1xPIII 800MHz / 512Mb RAM / IDE
      Result: 208.6
Host4: 1xPIII 600MHz / 256Mb RAM / IDE
      Result: 153.6

Host1:
Dhrystone 2 using register variables        116700.0  1139475.0       97.6
Double-Precision Whetstone                      55.0      295.2       53.7
Execl Throughput                                43.0      697.7      162.3
File Copy 1024 bufsize 2000 maxblocks         3960.0    83438.0      210.7
File Copy 256 bufsize 500 maxblocks           1655.0    38613.0      233.3
File Copy 4096 bufsize 8000 maxblocks         5800.0   121259.0      209.1
Pipe Throughput                              12440.0   254518.7      204.6
Process Creation                               126.0     1831.7      145.4
Shell Scripts (8 concurrent)                     6.0      201.7      336.2
System Call Overhead                         15000.0   251716.1      167.8
                                                                  =========
      FINAL SCORE                                                     164.7

Host2:
Dhrystone 2 using register variables        116700.0  1837005.3      157.4
Double-Precision Whetstone                      55.0      478.4       87.0
Execl Throughput                                43.0      923.8      214.8
File Copy 1024 bufsize 2000 maxblocks         3960.0    54663.0      138.0
File Copy 256 bufsize 500 maxblocks           1655.0    18601.0      112.4
File Copy 4096 bufsize 8000 maxblocks         5800.0    96807.0      166.9
Pipe Throughput                              12440.0   399294.5      321.0
Process Creation                               126.0     3503.8      278.1
Shell Scripts (8 concurrent)                     6.0      277.0      461.7
System Call Overhead                         15000.0   393738.7      262.5
                                                                  =========
      FINAL SCORE                                                     195.7

Host3:
Dhrystone 2 using register variables        116700.0  1506306.9      129.1
Double-Precision Whetstone                      55.0      364.4       66.3
Execl Throughput                                43.0     1103.3      256.6
File Copy 1024 bufsize 2000 maxblocks         3960.0   110403.0      278.8
File Copy 256 bufsize 500 maxblocks           1655.0    53065.0      320.6
File Copy 4096 bufsize 8000 maxblocks         5800.0   151107.0      260.5
Pipe Throughput                              12440.0   378679.6      304.4
Pipe-based Context Switching                  4000.0   161272.9      403.2
Process Creation                               126.0     7144.7      567.0
Shell Scripts (8 concurrent)                     6.0       20.0       33.3
System Call Overhead                         15000.0   412399.5      274.9
                                                                  =========
      FINAL SCORE                                                     208.6

Host4:
Dhrystone 2 using register variables        116700.0  1099714.7       94.2
Double-Precision Whetstone                      55.0      328.1       59.7
Execl Throughput                                43.0      692.1      161.0
File Copy 1024 bufsize 2000 maxblocks         3960.0    56029.0      141.5
File Copy 256 bufsize 500 maxblocks           1655.0    18783.0      113.5
File Copy 4096 bufsize 8000 maxblocks         5800.0    96243.0      165.9
Pipe Throughput                              12440.0   306228.7      246.2
Process Creation                               126.0     3937.3      312.5
Shell Scripts (8 concurrent)                     6.0      125.9      209.8
System Call Overhead                         15000.0   282470.2      188.3
                                                                  =========
      FINAL SCORE                                                     153.6

--
Jean-Eric Cuendet
Linkvest SA
Av des Baumettes 19, 1020 Renens Switzerland
Tel +41 21 632 9043  Fax +41 21 632 9090

http://www.linkvest.com
--------------------------------------------------------

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in

More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

 
 
 

UNIX bench better on 2.2 than 2.4?

Post by Jean-Eric Cuende » Sat, 16 Mar 2002 17:00:17


Sorry, I forgot to give kernel versions...

Hi,
I ran UNIX bench on 4 of our machines. The results are below.
What is strange is that 2.2 makes better than 2.4! Is it a known fact?
And is the UNIX bench better with 2 CPU?
-jec

Results:
Host1: 2xPIII 550MHz / 1Gb RAM / RAID5 SCSI / 2.4.6smp + LVM
       Result: 164.7
Host2: 2xPIII 866MHz / 1Gb RAM / RAID1 soft IDE / 2.4.16smp + LVM
       Result: 195.7
Host3: 1xPIII 800MHz / 512Mb RAM / IDE / 2.2.19 RedHat 6.2
       Result: 208.6
Host4: 1xPIII 600MHz / 256Mb RAM / IDE / 2.4.19-pre2-ac4-preempt
       Result: 153.6

Host1:
Dhrystone 2 using register variables        116700.0  1139475.0       97.6
Double-Precision Whetstone                      55.0      295.2       53.7
Execl Throughput                                43.0      697.7      162.3
File Copy 1024 bufsize 2000 maxblocks         3960.0    83438.0      210.7
File Copy 256 bufsize 500 maxblocks           1655.0    38613.0      233.3
File Copy 4096 bufsize 8000 maxblocks         5800.0   121259.0      209.1
Pipe Throughput                              12440.0   254518.7      204.6
Process Creation                               126.0     1831.7      145.4
Shell Scripts (8 concurrent)                     6.0      201.7      336.2
System Call Overhead                         15000.0   251716.1      167.8
                                                                   =========
       FINAL SCORE                                                     164.7

Host2:
Dhrystone 2 using register variables        116700.0  1837005.3      157.4
Double-Precision Whetstone                      55.0      478.4       87.0
Execl Throughput                                43.0      923.8      214.8
File Copy 1024 bufsize 2000 maxblocks         3960.0    54663.0      138.0
File Copy 256 bufsize 500 maxblocks           1655.0    18601.0      112.4
File Copy 4096 bufsize 8000 maxblocks         5800.0    96807.0      166.9
Pipe Throughput                              12440.0   399294.5      321.0
Process Creation                               126.0     3503.8      278.1
Shell Scripts (8 concurrent)                     6.0      277.0      461.7
System Call Overhead                         15000.0   393738.7      262.5
                                                                   =========
       FINAL SCORE                                                     195.7

Host3:
Dhrystone 2 using register variables        116700.0  1506306.9      129.1
Double-Precision Whetstone                      55.0      364.4       66.3
Execl Throughput                                43.0     1103.3      256.6
File Copy 1024 bufsize 2000 maxblocks         3960.0   110403.0      278.8
File Copy 256 bufsize 500 maxblocks           1655.0    53065.0      320.6
File Copy 4096 bufsize 8000 maxblocks         5800.0   151107.0      260.5
Pipe Throughput                              12440.0   378679.6      304.4
Pipe-based Context Switching                  4000.0   161272.9      403.2
Process Creation                               126.0     7144.7      567.0
Shell Scripts (8 concurrent)                     6.0       20.0       33.3
System Call Overhead                         15000.0   412399.5      274.9
                                                                   =========
       FINAL SCORE                                                     208.6

Host4:
Dhrystone 2 using register variables        116700.0  1099714.7       94.2
Double-Precision Whetstone                      55.0      328.1       59.7
Execl Throughput                                43.0      692.1      161.0
File Copy 1024 bufsize 2000 maxblocks         3960.0    56029.0      141.5
File Copy 256 bufsize 500 maxblocks           1655.0    18783.0      113.5
File Copy 4096 bufsize 8000 maxblocks         5800.0    96243.0      165.9
Pipe Throughput                              12440.0   306228.7      246.2
Process Creation                               126.0     3937.3      312.5
Shell Scripts (8 concurrent)                     6.0      125.9      209.8
System Call Overhead                         15000.0   282470.2      188.3
                                                                   =========
       FINAL SCORE                                                     153.6

--
Jean-Eric Cuendet
Linkvest SA
Av des Baumettes 19, 1020 Renens Switzerland
Tel +41 21 632 9043  Fax +41 21 632 9090

http://www.linkvest.com
--------------------------------------------------------

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in

More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

 
 
 

UNIX bench better on 2.2 than 2.4?

Post by Mark Mielk » Sat, 16 Mar 2002 19:40:12



> Results:
> Host1: 2xPIII 550MHz / 1Gb RAM / RAID5 SCSI / 2.4.6smp + LVM
>        Result: 164.7
> Host2: 2xPIII 866MHz / 1Gb RAM / RAID1 soft IDE / 2.4.16smp + LVM
>        Result: 195.7
> Host3: 1xPIII 800MHz / 512Mb RAM / IDE / 2.2.19 RedHat 6.2
>        Result: 208.6
> Host4: 1xPIII 600MHz / 256Mb RAM / IDE / 2.4.19-pre2-ac4-preempt
>        Result: 153.6

I dunno what Unix bench is... but isn't 153.6/208.6 close to 600/800
in terms of a fraction?

It isn't really fair to compare different machine configurations with
different OS's, and declare that the OS is the only factor involved.

Make Host3 or Host4 have both versions of the linux kernel in /boot
and try each on the same machine.

mark

--

.  .  _  ._  . .   .__    .  . ._. .__ .   . . .__  | Neighbourhood Coder
|\/| |_| |_| |/    |_     |\/|  |  |_  |   |/  |_   |
|  | | | | \ | \   |__ .  |  | .|. |__ |__ | \ |__  | Ottawa, Ontario, Canada

  One ring to rule them all, one ring to find them, one ring to bring them all
                       and in the darkness bind them...

                           http://mark.mielke.cc/

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in

More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

 
 
 

UNIX bench better on 2.2 than 2.4?

Post by Jean-Eric Cuende » Sat, 16 Mar 2002 19:50:14


Quote:

>>Results:
>>Host1: 2xPIII 550MHz / 1Gb RAM / RAID5 SCSI / 2.4.6smp + LVM
>>       Result: 164.7
>>Host2: 2xPIII 866MHz / 1Gb RAM / RAID1 soft IDE / 2.4.16smp + LVM
>>       Result: 195.7
>>Host3: 1xPIII 800MHz / 512Mb RAM / IDE / 2.2.19 RedHat 6.2
>>       Result: 208.6
>>Host4: 1xPIII 600MHz / 256Mb RAM / IDE / 2.4.19-pre2-ac4-preempt
>>       Result: 153.6

>I dunno what Unix bench is... but isn't 153.6/208.6 close to 600/800
>in terms of a fraction?

Yes, you are right for this one.
But what impressed me is between Host2/Host3. One is a dual-866 with
2.4.16 and the other is mono-800 with 2.2.19. And the host3 with 2.2.19
is ~7% better than Host2 (Not counting that Host3 has ~8% more CPU
clock, which should have be ~15% better for Host2 vs Host3...)

Quote:>Make Host3 or Host4 have both versions of the linux kernel in /boot
>and try each on the same machine.

But that's just what I'd like to see. How 2.4 compares vs 2.2
-jec

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in

More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

 
 
 

UNIX bench better on 2.2 than 2.4?

Post by Mark Mielk » Sat, 16 Mar 2002 20:10:14



> Yes, you are right for this one.
> But what impressed me is between Host2/Host3. One is a dual-866 with
> 2.4.16 and the other is mono-800 with 2.2.19. And the host3 with 2.2.19
> is ~7% better than Host2 (Not counting that Host3 has ~8% more CPU
> clock, which should have be ~15% better for Host2 vs Host3...)

There is no guarantee that UNIX bench is SMP-friendly, or even
designed to test an SMP machine. Single programs that make good use of
an SMP machine usually have to be designed very differently than a
traditional program designed for a single processor.

It is probably just luck that the numbers jump around.

Just because Unix bench performs better or worse, or not as better, on
an SMP machine, as a non-SMP machine, does not mean that the numbers
accurately effect the capabilities of a real-life application. Unless
specifically designed for the environment, it is even possible that the
numbers would not even come close.

mark

--

.  .  _  ._  . .   .__    .  . ._. .__ .   . . .__  | Neighbourhood Coder
|\/| |_| |_| |/    |_     |\/|  |  |_  |   |/  |_   |
|  | | | | \ | \   |__ .  |  | .|. |__ |__ | \ |__  | Ottawa, Ontario, Canada

  One ring to rule them all, one ring to find them, one ring to bring them all
                       and in the darkness bind them...

                           http://mark.mielke.cc/

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in

More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

 
 
 

1. Upgrading kernel 2.2.x to 2.4.x and GLibc 2.1.3 to 2.2.x

I am running SuSE6.4/kernel-2.2.14/GLIBC-2.1.3

I do not dare to upgrade GLIBC to 2.2.x since I could not find the
instructions/requirements to do be certain that it is safe.

I do not dare to install kernel 2.4.x for the same reasons.

Can somebody instruct me where to find information ?

Iztok

2. Jumbo Desktop

3. AIM Performance Bench results of both 2.4.xx and 2.5.xx kernels

4. Module Development Question

5. TIO bench mm2 patch better than mm1 patch

6. MIGRATION AIX

7. gdb on 2.2/2.4

8. grep, help

9. MSS in 2.2 -> 2.4

10. net.ipv4.tcp differences from 2.2 to 2.4.x kernel

11. OOM Killers in 2.2.x and 2.4.x?

12. 2.0, 2.2, 2.4, 2.5: fsync buffer race

13. Firewall - migrating from 2.2 kernel to 2.4