doubt about iostat output

Charles T. Bell cbell44 at cfl.rr.com
Wed May 25 20:14:53 UTC 2016


I would say that the write speed is faster, but that does not mean
that it is writing for the same period of time that the other
disk is being read.  Writing to a "clean" drive means that it can
write at the faster speed, while the other drive is reading at a
slower speed.  If it finished writing before the other disk finished
reading, then I would say something is wrong.
Good luck!

Tom

On 05/25/2016 06:43 AM, Luca Ferrari wrote:
> Hi all,
> I'm copying data between two USB disk, both ext4. Now the source hdd
> provides 1MB/s throughput, as reported by hdparm tests, but the dest
> hdd is writing at an higher speed. The thing is that monitoring with
> iostat the copy I _always_ see at least a double speed of writing and
> of transactions on the destination hdd. For instance:
> 
> avg-cpu:  %user   %nice %system %iowait  %steal   %idle
>            3,52    0,00    1,01   50,00    0,00   45,48
> 
> Device:            tps    kB_read/s    kB_wrtn/s    kB_read    kB_wrtn
> 
> sdc              26,50         0,00      2970,00          0       5940
> sdd              10,00      1024,00         0,00       2048          0
> 
> My doubt is: if the source hdd (sdd) is doing 10 tps), how can the
> destination hdd (sdc) _always_ write the double? I don't get how and
> where the data read is cached, since also the destination tps are
> higher than the source one.
> 
> Thanks,
> Luca
> 

-- 
"How come when it's us, it's an abortion, and
when it's a chicken, it's an omelette?" - George Carlin
-------------- next part --------------
A non-text attachment was scrubbed...
Name: cbell44.vcf
Type: text/x-vcard
Size: 149 bytes
Desc: not available
URL: <https://lists.ubuntu.com/archives/kubuntu-users/attachments/20160525/32e0e353/attachment.vcf>


More information about the kubuntu-users mailing list