Re: asymmetric NFS transfer rates

From: Emanuel Strobl <Emanuel.Strobl_at_gmx.net>
Date: Tue, 2 Nov 2004 19:05:01 +0100
Am Dienstag, 2. November 2004 17:50 schrieb Andre Guibert de Bruet:
> On Tue, 2 Nov 2004, Emanuel Strobl wrote:
> > Am Dienstag, 2. November 2004 13:47 schrieb Claus Guttesen:
> >>> problems, but I'm wondering why I
> >>> can't write to my 5.3-stable NFS server more that
> >>> 3,5MB/s while reading gives
> >>> me 9,5MB/s?
> >>
> >> Are you using IDE- or SCSI-disks? Reading is faster
> >> than writing on IDE.
> >
> > Like I wrote, the server writes more than 35MB/s onto the RAID5 array.
> > Last time I saw a hard drive which has problems with 10MB/s was 6 years
> > ago, no matter of IDE or SCSI.
>
> You are going by the assumption that the benchmarks are run using
> transfers of a single, contiguous, abnormally huge file. In the real
> world, things aren't that peachy. Remember to factor in things such as
> seek times, drive response latency, driver locking and contention on
> Giant (This is a SCSI RAID card, isn't it?), among other things...

It's a IDE Raid controller (3ware 7506-4, a real one) and the file is indeed 
huge, but not abnormally. I have a harddisk video recorder, so I have lots of 
700MB files. Also if I copy my photo collection from the server it takes 5 
Minutes but copying _to_ the server it takes almost 15 Minutes and the 
average file size is 5 MB. Fast Ethernet isn't really suitable for my needs, 
but at least the 10MB/s should be reached. I can't imagine I get better 
speeds when I upgrade to GbE, (which the important boxes are already, just 
not the switch) because NFS in it's current state isn't able to saturate a 
100baseTX line, at least in one direction. That's the real anstonishing thing 
for me. Why does reading staurate 100BaseTX but writes only a third?

Thanks,

-Mano

>
> Regards,
> Andy
>
> | Andre Guibert de Bruet | Enterprise Software Consultant >
> | Silicon Landmark, LLC. | http://siliconlandmark.com/    >
>
> _______________________________________________
> freebsd-current_at_freebsd.org mailing list
> http://lists.freebsd.org/mailman/listinfo/freebsd-current
> To unsubscribe, send any mail to "freebsd-current-unsubscribe_at_freebsd.org"

Received on Tue Nov 02 2004 - 17:05:20 UTC

This archive was generated by hypermail 2.4.0 : Wed May 19 2021 - 11:38:20 UTC