Re: libata - anomaly performance with NCQ

[Bartosz SKOWRON]

On Sat, Jan 31, 2009 at 6:18 AM, Robert Hancock <hancockrwd@xxxxxxxxx> wrote:

> bonnie++ is one I've used, I'm sure there are others.. Of course, what do
> you compare the results to..

Hmmm more tests == more confusions ;-)

NCQ ENABLED:

Version 1.03b       ------Sequential Output------ --Sequential Input- --Random-
                    -Per Chr- --Block-- -Rewrite- -Per Chr- --Block-- --Seeks--
Machine        Size K/sec %CP K/sec %CP K/sec %CP K/sec %CP K/sec %CP  /sec %CP
banan            6G 32681  99 81220  35 37915  31 32675  99 88157  28 286.0   0
                    ------Sequential Create------ --------Random Create--------
                    -Create-- --Read--- -Delete-- -Create-- --Read--- -Delete--
              files  /sec %CP  /sec %CP  /sec %CP  /sec %CP  /sec %CP  /sec %CP
                 16  3115  32 +++++ +++  3774  38  3321  38 +++++ +++  1004  10

NCQ DISABLED:
Version 1.03b       ------Sequential Output------ --Sequential Input- --Random-
                    -Per Chr- --Block-- -Rewrite- -Per Chr- --Block-- --Seeks--
Machine        Size K/sec %CP K/sec %CP K/sec %CP K/sec %CP K/sec %CP  /sec %CP
banan            6G 32617  99 84305  36 36022  16 32889  99 88199  28 233.9   0
                    ------Sequential Create------ --------Random Create--------
                    -Create-- --Read--- -Delete-- -Create-- --Read--- -Delete--
              files  /sec %CP  /sec %CP  /sec %CP  /sec %CP  /sec %CP  /sec %CP
                 16  3866  40 +++++ +++  5538  48  3668  38 +++++ +++  1093  11


So it looks very similar but without ncq looks a bit better, doesn't it?

something practical test: copy a movie file (~750MB) into /tmp (tmpfs):
ncq: ~70-75MB/s
noncq: ~65-70MB/s

and hdparm once again:

ncq:
 Timing buffered disk reads:  108 MB in  3.06 seconds =  35.33 MB/sec
noncq:
 Timing buffered disk reads:  258 MB in  3.01 seconds =  85.63 MB/sec


Can you or anyone else analyze those results? They confuse me a lot
cause they are sometimes in a conflict.
And what results of bonnie++ should i compare to other tests?
sequintal output:block ? or per?

and here is my hdd:
 Model=ST9320421AS                             , FwRev=SD13    ,
SerialNo=            5TJ0DGSP
 Config={ HardSect NotMFM HdSw>15uSec Fixed DTR>10Mbs RotSpdTol>.5% }
 RawCHS=16383/16/63, TrkSize=0, SectSize=0, ECCbytes=4
 BuffType=unknown, BuffSize=16384kB, MaxMultSect=16, MultSect=?16?
 CurCHS=16383/16/63, CurSects=16514064, LBA=yes, LBAsects=625142448
 IORDY=on/off, tPIO={min:120,w/IORDY:120}, tDMA={min:120,rec:120}
 PIO modes:  pio0 pio1 pio2 pio3 pio4
 DMA modes:  mdma0 mdma1 mdma2
 UDMA modes: udma0 udma1 udma2 udma3 udma4 udma5 *udma6
 AdvancedPM=yes: unknown setting WriteCache=enabled
 Drive conforms to: unknown:  ATA/ATAPI-4,5,6,7


regards,
bart
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/