WimMark I for 2.5.64-mm1

WimMark I for 2.5.64-mm1

Post by Joel Becke » Sat, 08 Mar 2003 20:10:04



WimMark I report for 2.5.64-mm1

Runs with anticipatory scheduler:  547.28 580.69
Runs with deadline scheduler:  1557.79 1360.52

        WimMark I is a rough benchmark we have been running
here at Oracle against various kernels.  Each run tests an OLTP
workload on the Oracle database with somewhat restrictive memory
conditions.  This reduces in-memory buffering of data, allowing for
more I/O.  The I/O is read and sync write, random and seek-laden.
        The benchmark is called "WimMark I" because it has no
official standing and is only a relative benchmark useful for comparing
kernel changes.  The benchmark is normalized an arbitrary kernel, which
scores 1000.0.  All other numbers are relative to this.
        The machine in question is a 4 way 700 MHz Xeon machine with 2GB
of RAM.  CONFIG_HIGHMEM4GB is selected.  The disk accessed for data is a
10K RPM U2W SCSI of similar vintage.  Unless mentioned, all runs are
on this machine (variation in hardware would indeed change the
benchmark).

--

"The lawgiver, of all beings, most owes the law allegiance.  He of all
 men should behave as though the law compelled him.  But it is the
 universal weakness of mankind that what we are given to administer we
 presently imagine we own."
        - H.G. Wells

Joel Becker
Senior Member of Technical Staff
Oracle Corporation

Phone: (650) 506-8127
-
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/

 
 
 

WimMark I for 2.5.64-mm1

Post by Alex Riese » Sat, 08 Mar 2003 20:10:22


Joel Becker, Fri, Mar 07, 2003 18:57:01 +0100:

Quote:

> WimMark I report for 2.5.64-mm1

> Runs with anticipatory scheduler:  547.28 580.69
> Runs with deadline scheduler:  1557.79 1360.52

What do the numbers mean?
Is AS better or worse DS?

Quote:>    WimMark I is a rough benchmark we have been running
> here at Oracle against various kernels.  Each run tests an OLTP
> workload on the Oracle database with somewhat restrictive memory
> conditions.  This reduces in-memory buffering of data, allowing for
> more I/O.  The I/O is read and sync write, random and seek-laden.
>    The benchmark is called "WimMark I" because it has no
> official standing and is only a relative benchmark useful for comparing
> kernel changes.  The benchmark is normalized an arbitrary kernel, which
> scores 1000.0.  All other numbers are relative to this.
>    The machine in question is a 4 way 700 MHz Xeon machine with 2GB
> of RAM.  CONFIG_HIGHMEM4GB is selected.  The disk accessed for data is a
> 10K RPM U2W SCSI of similar vintage.  Unless mentioned, all runs are
> on this machine (variation in hardware would indeed change the
> benchmark).

-
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/

 
 
 

WimMark I for 2.5.64-mm1

Post by Andrew Morto » Sat, 08 Mar 2003 20:50:11



> WimMark I report for 2.5.64-mm1

> Runs with anticipatory scheduler:  547.28 580.69
> Runs with deadline scheduler:  1557.79 1360.52

Boggle.

I have a patch in my inbox which increases NickMark I throughput by 400%, so
it should help this one.

Is the difference between 2.5.64 and 2.5.64-mm1 statistically significant?
(It should be - the readahead changes).

-
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/

 
 
 

WimMark I for 2.5.64-mm1

Post by Joel Becke » Sat, 08 Mar 2003 21:20:18



> > Runs with anticipatory scheduler:  547.28 580.69
> > Runs with deadline scheduler:  1557.79 1360.52

> What do the numbers mean?
> Is AS better or worse DS?

        This is an OLTP setup, so the numbers are representative of
transactional load.  Bigger is better.

Joel

--

"The suffering man ought really to consume his own smoke; there is no
 good in emitting smoke till you have made it into fire."
                        - thomas carlyle

Joel Becker
Senior Member of Technical Staff
Oracle Corporation

Phone: (650) 506-8127
-
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/

 
 
 

WimMark I for 2.5.64-mm1

Post by Joel Becke » Sat, 08 Mar 2003 21:30:11



> Is the difference between 2.5.64 and 2.5.64-mm1 statistically significant?
> (It should be - the readahead changes).

        Yes, it is.  The average is noticeably different, and the high
number for 2.5.64 isn't much higher than the low number for 2.5.64-mm1.

Joel

--

Life's Little Instruction Book #157

        "Take time to smell the roses."

Joel Becker
Senior Member of Technical Staff
Oracle Corporation

Phone: (650) 506-8127
-
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/

 
 
 

WimMark I for 2.5.64-mm1

Post by Randy.Dunla » Sat, 08 Mar 2003 21:40:07



| WimMark I report for 2.5.64-mm1
|
| Runs with anticipatory scheduler:  547.28 580.69
| Runs with deadline scheduler:  1557.79 1360.52
|
|       WimMark I is a rough benchmark we have been running
| here at Oracle against various kernels.  Each run tests an OLTP
| workload on the Oracle database with somewhat restrictive memory
| conditions.  This reduces in-memory buffering of data, allowing for
| more I/O.  The I/O is read and sync write, random and seek-laden.
|       The benchmark is called "WimMark I" because it has no
| official standing and is only a relative benchmark useful for comparing
| kernel changes.  The benchmark is normalized an arbitrary kernel, which
| scores 1000.0.  All other numbers are relative to this.
|       The machine in question is a 4 way 700 MHz Xeon machine with 2GB
| of RAM.  CONFIG_HIGHMEM4GB is selected.  The disk accessed for data is a
| 10K RPM U2W SCSI of similar vintage.  Unless mentioned, all runs are
| on this machine (variation in hardware would indeed change the
| benchmark).

Is there a web page where we can view/compare results?

Thanks,
--
~Randy
-
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/

 
 
 

WimMark I for 2.5.64-mm1

Post by Joel Becke » Sat, 08 Mar 2003 21:50:13



> Is there a web page where we can view/compare results?

        No, but that's a good idea!  When I have one, I'll post it.

Joel

--

"Reality is merely an illusion, albeit a very persistent one."
        - Albert Einstien

Joel Becker
Senior Member of Technical Staff
Oracle Corporation

Phone: (650) 506-8127
-
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. WimMark I for 2.5.64-mm4

WimMark I report for 2.5.64-mm4

Runs with anticipatory scheduler:  1066.70 1291.23
Runs with deadline scheduler:  1655.77 1594.73 1610.91

        WimMark I is a rough benchmark we have been running
here at Oracle against various kernels.  Each run tests an OLTP
workload on the Oracle database with somewhat restrictive memory
conditions.  This reduces in-memory buffering of data, allowing for
more I/O.  The I/O is read and sync write, random and seek-laden.
        The benchmark is called "WimMark I" because it has no
official standing and is only a relative benchmark useful for comparing
kernel changes.  The benchmark is normalized an arbitrary kernel, which
scores 1000.0.  All other numbers are relative to this.
        The machine in question is a 4 way 700 MHz Xeon machine with 2GB
of RAM.  CONFIG_HIGHMEM4GB is selected.  The disk accessed for data is a
10K RPM U2W SCSI of similar vintage.  The data files are living on an
ext3 filesystem.  Unless mentioned, all runs are
on this machine (variation in hardware would indeed change the
benchmark).

--

"You cannot bring about prosperity by discouraging thrift. You cannot
 strengthen the weak by weakening the strong. You cannot help the wage
 earner by pulling down the wage payer. You cannot further the
 brotherhood of man by encouraging class hatred. You cannot help the
 poor by destroying the rich. You cannot build character and courage by
 taking away a man's initiative and independence. You cannot help men
 permanently by doing for them what they could and should do for themselves."
        - Abraham Lincoln

Joel Becker
Senior Member of Technical Staff
Oracle Corporation

Phone: (650) 506-8127
-
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/

2. Looking for a simple serial I/O program

3. WimMark I for 2.5.64-mm2

4. Minimizing the number of "setuid root" daemons

5. WimMark I for 2.5.64

6. AVI's in linux

7. [2.5.64] kexec for 2.5.64 available

8. DIP/SLIP +16 Connections (SL_SLIP_LOTS)

9. 2.5.64-mm1: Badness in request_irq

10. 2.5.64-mm1 with contest

11. 2.5.64-mm1

12. Hard lock with Kernel 2.5.64-2.5.65-mm1 starting XFree 4.3.0