-=>Viper<=-
On Sun, 31 Oct 1999, Brian Craft wrote:
> Hi all -- I was recently trying to record a high bandwidth signal & ran
> into the bandwidth limit of my ultra-dma drive.
>
> I've hacked the program & also bonnie-big to write alternate blocks to two
> files on different ide interfaces, with the idea that I'd get more bandwidth
> writing to two disks.
>
> I get the odd result that this actually slows down the data rate.
>
> Anyone know what's going on here? Is there some other bottleneck? Is
> software raid going to give me the same result (slower than one drive)?
>
> I'm running an ASUS P2B-D, 2.2 kernel. hdparm for both drives are set to
> "-d 1 -m 16 -u 1 -c 1".
>
>
>
> This result is basically the same for either drive. One is primary master,
> the other secondary slave. The secondary master is a cd burner which is
> not accessed during the test:
>
> -------Sequential Output-------- ---Sequential Input-- --Random--
> -Per Char- --Block--- -Rewrite-- -Per Char- --Block--- --Seeks---
> Machine MB K/sec %CPU K/sec %CPU K/sec %CPU K/sec %CPU K/sec %CPU /sec %CPU
> 1* 600 5402 79.8 10464 14.7 4073 10.7 6708 94.8 11492 12.5 83.0 1.7
>
>
> This is the result for writing to two disks:
>
> -------Sequential Output-------- ---Sequential Input-- --Random--
> -Per Char- --Block--- -Rewrite-- -Per Char- --Block--- --Seeks---
> Machine MB K/sec %CPU K/sec %CPU K/sec %CPU K/sec %CPU K/sec %CPU /sec %CPU
> 2* 300 5355 79.4 9913 13.7 4058 10.4 6218 85.5 10983 12.4 70.8 1.9
>
>
> b.c.
>
> -
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@vger.rutgers.edu
> Please read the FAQ at http://www.tux.org/lkml/
>
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/