WimMark I for 2.5.64-mm1

WimMark I for 2.5.64-mm1

[Joel Becker]

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
E-mail: joel.becker@oracle.com
Phone: (650) 506-8127

Re: WimMark I for 2.5.64-mm1

[Alex Riesen]

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

> 	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).

Re: WimMark I for 2.5.64-mm1

[Andrew Morton]

Joel Becker <Joel.Becker@oracle.com> wrote:
>
> 
> 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).

Re: WimMark I for 2.5.64-mm1

[Joel Becker]

On Fri, Mar 07, 2003 at 07:06:53PM +0100, Alex Riesen wrote:
> > 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
E-mail: joel.becker@oracle.com
Phone: (650) 506-8127

Re: WimMark I for 2.5.64-mm1

[Joel Becker]

On Fri, Mar 07, 2003 at 10:43:33AM -0800, Andrew Morton wrote:
> 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
E-mail: joel.becker@oracle.com
Phone: (650) 506-8127

Re: WimMark I for 2.5.64-mm1

[Randy.Dunlap]

On Fri, 7 Mar 2003 09:57:01 -0800 Joel Becker <Joel.Becker@oracle.com> wrote:

| 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

Re: WimMark I for 2.5.64-mm1

[Joel Becker]

On Fri, Mar 07, 2003 at 11:28:40AM -0800, Randy.Dunlap wrote:
> 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
E-mail: joel.becker@oracle.com
Phone: (650) 506-8127