Re: IDE drive data transfer rates, kernel optimization?

[Geert Uytterhoeven]

On Wed, 5 Aug 1998, Mike A. Harris wrote:
> I am concerned about the advertised data rates of IDE hard disks,
> more specifically the "true maximum throughput" of said drives.
> 
> I have a Quantum Fireball SE 4.3G drive that claims UDMA
> capability.  I am getting 7.8Mb/s out of this drive.  This is
> fast, and I can't complain, however it is nowhere near the rated
> capacity of the drive which is 33Mb/s on the SE pdf datasheet.

The 33 MB/s is the maximum transfer rate from buffer to host.

> Further inspection of the datasheet indicates that the drive has
> an internal buffer size of 128k.  When I do an "hdparm -i" on the
> drive, it indicates:

Yes, modern drives have too small buffers, unless they are expensive.

> 1 root@red:/# hdparm -i /dev/hda
> 
> /dev/hda:
> 
>  Model=QUANTUM FIREBALL SE4.3A, FwRev=API.0A00, SerialNo=33473491
>  Config={ HardSect NotMFM HdSw>15uSec Fixed DTR>10Mbs }
>  RawCHS=14848/9/63, TrkSize=32256, SectSize=512, ECCbytes=4
>  BuffType=3(DualPortCache), BuffSize=80kB, MaxMultSect=16, MultSect=16
>  DblWordIO=no, maxPIO=2(fast), DMA=yes, maxDMA=2(fast)
>  CurCHS=14848/9/63, CurSects=8418816, LBA=yes, LBAsects=8418816
>  tDMA={min:120,rec:120}, DMA modes: sword0 sword1 sword2 mword0 mword1 mword2
>  IORDY=on/off, tPIO={min:120,w/IORDY:120}, PIO modes: mode3 mode4
> 
> 
> As you can see, it says "BuffSize=80kB".  That is NOT 128k for
> sure.

Modern drives have very small ROMs, and load parts of their Firmware from disk
into the buffer. Yes, this makes the small buffer even smaller.

> Some other interesting things... the pdf datasheet says that the
> drive has an internal data transfer rate (disk to buffer) of 
> 158Mb/s.  Other parts of the sheet refer to MEGABYTE as MB, so
> I'm assuming that Mb means megabit which makes more sense than
> does Megabyte for the number "158".  So at 158 megabits per
> second data transfer from disk to buffer, we get:
> 
>  158 / 8 = 19.75MB/s  or 19 megabytes per second.
> 
> This is NOT the rated 33megs per second that the sheet claims for
> UDMA operation.  The rate for PIO/DMA is 16.6Mb/s.  I get 7.8Mb/s

No, 33 MB/s is buffer to host only. And due to the small buffer, you can
sustain that rate for a fraction of a second only.

> What gives with these specs?  What is even more confusing is that
> at the bottom of the document it says "Quantum defines a megabyte
> as 1000000 bytes".  That adds a hell of a lot of confusion to
> their datasheets, and accepted standards.  I realize that Quantum
> is not the only drive manufacturer doing this however - they all
> do it.

Allmost all (all?) drive manufacturers define 1 MB = 1E6 bytes. This used
to make a small difference in the 20 MB era, but for a `4.3' GB disk, the
difference is 300 MB.

Summarized: don't believe marketing. There is a reason why expensive SCSI disks
are expensive.

Greetings,

						Geert

--
Geert Uytterhoeven                     [e-mail]
Wavelets, Linux/{m68k~Amiga,PPC~CHRP}  http://www.cs.kuleuven.ac.be/~geert/
Department of Computer Science -- Katholieke Universiteit Leuven -- Belgium