Re: fsync on large files

kwrohrer@ce.mediaone.net
Wed, 17 Feb 1999 18:03:49 -0600 (EST)

And lo, Dan Hollis saith unto me:
> On Wed, 17 Feb 1999, Alan Cox wrote:
> > > I've also destroyed maxtors with synchronous syslog. I dont think any
> > > drive is up to the task of 50 writes/sec on the same spot on the disk 24/7
> > > for weeks on end.
> > According to the vendors own claims it is. From my own experience the
> > ibm drives seem to take it fine, and worse.
>
> Write a program which sync writes to the same spot 50 times/sec 24/7 and
> get back to me in two months or so if your ibm drive is still alive.

"The" same spot? Wouldn't that be the same *two* spots, presumably with a
seek involved? *That* would induce a 50 Hz vibration which could be
nasty. But just writing to one sector 50 times a second should be an
absolute fiesta of...rotational latency. Doldrum city. Silence and
stillness, at least from the head mechanism...


> Thats 129,600,000 writes to the same sector on the disk each month.

Now, 129,600,000 round trips between the end of the log file and the
log file inode could cause a bit of wear...

Keith 

-- 
 "Well, look at that.  The sun's   | Linux: http://www.linuxhq.com     |"Zooty,
  coming up." -- John Sheridan,    | KDE:   http://www.kde.org         | zoot
  "Sleeping in Light", Babylon 5   | Keith: kwrohrer@enteract.com      | zoot!"
www.midwinter.com/lurk/lurker.html | http://www.enteract.com/~kwrohrer | --Rebo


-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/