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

RE: Re: IDE hdd faster than sata? How come?



Run each test 10 times and take the average.
Secondly, is it attached to your motherboard (SATA) or a PCI card?

Justin.

-----Original Message-----
From: news [mailto:news@sea.gmane.org] On Behalf Of Hugo Vanwoerkom
Sent: Tuesday, October 25, 2005 6:57 AM
To: debian-user@lists.debian.org
Subject: Re: IDE hdd faster than sata? How come?

Bruno Buys wrote:
> How good is hdparm benchmark for sata? What am I missing here?
> 
> 
> frank:/home/bruno# hdparm -Tt /dev/sda
> 
> /dev/sda:
> Timing cached reads:   2368 MB in  2.00 seconds = 1183.00 MB/sec
> Timing buffered disk reads:  170 MB in  3.01 seconds =  56.56 MB/sec
> 
> 
> frank:/home/bruno# hdparm -Tt /dev/hda
> 
> /dev/hda:
> Timing cached reads:   2352 MB in  2.00 seconds = 1176.18 MB/sec
> Timing buffered disk reads:  180 MB in  3.03 seconds =  59.47 MB/sec
> 
> 
> IDE disk is SAMSUNG SP0802N, FwRev=TK200-04. 80GB. udma5.
> Sata is SAMSUNG SP0812C SATA 80GB.
> 
> 
> The only thing I found was in samsung website "The drive comes
defaulted 
> to udma100. To enable udma133 refer to www.samsung.com.br". So far, 
> www.samsung.com.br gives me nothing.
> 

No clue.
but I have exactly the same ATA Samsung disk.
That gives:

  Timing cached reads:   1348 MB in  2.01 seconds = 671.75 MB/sec
  Timing buffered disk reads:  164 MB in  3.00 seconds =  54.64 MB/sec

Same udma5.
But smartctl -A -v 194,unknown /dev/hdc gives:
...
199 UDMA_CRC_Error_Count    0x003e   200   195   000    Old_age   Always

       -       92
...

And a few days ago when running Sarge on one of its partitions I got all

sorts of trouble:
...
kernel: Buffer I/O error on device hdb3, logical block 163858
kernel: lost page write due to I/O error on hdb3
...
kernel: EXT2-fs error (device hdb3): ext2_readdir: bad page in #83892
kernel: Remounting filesystem read-only
...

That disk is 6 months old!
So I am going to get rid of the Samsung and get something else.

H







-- 
To UNSUBSCRIBE, email to debian-user-REQUEST@lists.debian.org 
with a subject of "unsubscribe". Trouble? Contact
listmaster@lists.debian.org



Reply to: