[mythtv-users] RAID5 ATA IDE hardware card recommendations formythbackend machine

Yan-Fa Li yanfali at best.com
Fri Dec 10 07:00:42 UTC 2004

I am evil :D

John Andersen wrote:
> Continuing with this evil top posting.....
> Your analogy is largely obsolete, as drive manufactures took steps
> to utilize the the surface uniformly years ago.
> On Wednesday 08 December 2004 16:30, Yan-Fa Li wrote:
>>Think of an old fashioned 78RPM record player.  The disc spins at a
>>constant speed, 78RPM.  The outer part of the disc is moving faster than
>>the inner part, but the time the head has to read and write is the same
>>regardless of where you are on the disc.  Therefore the inner part gets
>>data faster because it has to spin much less to read the same amount of
>>data assuming all the tracks hold exactly the same amount.

I agree with you that the mapping information is now handled 
differently, i.e. that when I say track 80, sector 50, it's remapped 
internally to something different, but the principle of the example 
still hold true:


Also if I do hdpart -t on each zone of a pair of 160GB hard disks I 

[root at mythbackend root]# hdparm -t /dev/md1

  Timing buffered disk reads:  136 MB in  3.01 seconds =  45.18 MB/sec
[root at mythbackend root]# hdparm -t /dev/md2

  Timing buffered disk reads:  126 MB in  3.01 seconds =  41.86 MB/sec
[root at mythbackend root]# hdparm -t /dev/md3

  Timing buffered disk reads:  116 MB in  3.03 seconds =  38.28 MB/sec
[root at mythbackend root]# hdparm -t /dev/md4

  Timing buffered disk reads:   98 MB in  3.01 seconds =  32.56 MB/sec

Empirically that demonstrates that the principle is still valid.  Inner 
parts of the disk perform at higher rates than the outer part.  The 
result is valid on both the RAID partitions and the hard drives directly.

[root at mythbackend root]# cat /proc/mdstat
Personalities : [raid1]
read_ahead 1024 sectors
md4 : active raid1 hdi4[0] hdk4[1]
       40098176 blocks [2/2] [UU]

md3 : active raid1 hdi3[0] hdk3[1]
       39993728 blocks [2/2] [UU]

md2 : active raid1 hdi2[0] hdk2[1]
       39993728 blocks [2/2] [UU]

md1 : active raid1 hdi1[0] hdk1[1]
       39993664 blocks [2/2] [UU]

[root at mythbackend root]# hdparm -i /dev/hdi


  Model=Maxtor 4R160L0, FwRev=RAMB1TU0, SerialNo=R4202E2E
  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=268435455
  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: disabled (255) WriteCache=enabled
  Drive conforms to: (null):

  * signifies the current active mode

Try this experiment for yourself.  You'll see what I mean.


More information about the mythtv-users mailing list