[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Schedulers



Here ya go kiddies, I ran hdparm -t -T at init 1(single user) while
running deadline and rts(realtime scheduler) schedulers.  In these tests
the rts performed much better than deadline.  However, these tests were
only reads(which is what the rts is optimized for).  For those of you
who don't know, when running rts(the default I/O scheduler) when the
system makes a read it waits 2 ms to anticipate another read.  The
deadline I/O scheduler has all requests in a queue and makes sure that
no request waits too long.  I have found that in actual operation the
deadline scheduler performs much better.  Seldom is the computer *JUST*
reading disk I/O, plus if you are requesting reads from multiple files
the rts loses again.  I have found that under load( running X Windows
System,gaim, amsn, xchat, evolution, xmms, XFCE, rox-filer, Mozilla
Firebird, bunning a cd, and updating the system(apt)) the deadline
scheduler is very responsive, far more than the rts under similar load
conditions.

-Michael Moran

PS sorry about the rts only having one trial but i forgot to append the
-a flag on tee after i rebooted, and i don't feel like rebooting again
though the data was all within .3 MB/s, though I found that after
loading programs and then going to init 1 the results can be rather
differnet(though everything should be flushed) another test another day
:)


#data follows

#deadline
/dev/hda:
Timing buffer-cache reads:   624 MB in  2.01 seconds = 310.96 MB/sec
Timing buffered disk reads:   96 MB in  3.04 seconds =  31.54 MB/sec

/dev/hda:
Timing buffer-cache reads:   624 MB in  2.01 seconds = 310.50 MB/sec
Timing buffered disk reads:   96 MB in  3.04 seconds =  31.55 MB/sec

/dev/hdb:
Timing buffer-cache reads:   624 MB in  2.01 seconds = 310.96 MB/sec
Timing buffered disk reads:  120 MB in  3.02 seconds =  39.69 MB/sec

/dev/hdb:
Timing buffer-cache reads:   624 MB in  2.01 seconds = 310.80 MB/sec
Timing buffered disk reads:  120 MB in  3.02 seconds =  39.79 MB/sec

#rts


/dev/hda:
 Timing buffer-cache reads:   688 MB in  2.01 seconds = 342.34 MB/sec
 Timing buffered disk reads:   94 MB in  3.01 seconds =  31.22 MB/sec


/dev/hdb:
 Timing buffer-cache reads:   668 MB in  2.00 seconds = 333.22 MB/sec
 Timing buffered disk reads:  120 MB in  3.02 seconds =  39.74 MB/sec


# Drive specs
/dev/hda:

Model=Maxtor 53073H4, FwRev=JAC61HU0, SerialNo=F40HEDBC
Config={ Fixed }
RawCHS=16383/16/63, TrkSize=0, SectSize=0, ECCbytes=57
BuffType=DualPortCache, BuffSize=2048kB, MaxMultSect=16, MultSect=16
CurCHS=16383/16/63, CurSects=16514064, LBA=yes, LBAsects=60030432
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 
AdvancedPM=yes: disabled (255) WriteCache=enabled
Drive conforms to: ATA/ATAPI-6 T13 1410D revision 0: 

* signifies the current active mode


/dev/hdb:

Model=MAXTOR 6L060J3, FwRev=A93.0500, SerialNo=663212116846
Config={ HardSect NotMFM HdSw>15uSec Fixed DTR>10Mbs }
RawCHS=16383/16/63, TrkSize=32256, SectSize=21298, ECCbytes=4
BuffType=DualPortCache, BuffSize=1819kB, MaxMultSect=16, MultSect=16
CurCHS=16383/16/63, CurSects=16514064, LBA=yes, LBAsects=117266688
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=no WriteCache=enabled
Drive conforms to: ATA/ATAPI-5 T13 1321D revision 1: 

* signifies the current active mode