Re: (reiserfs) Re: LVM / Filesystems / High availability

Albert D. Cahalan (acahalan@cs.uml.edu)
Sat, 27 Jun 1998 04:54:20 -0400 (EDT)


Stephen C. Tweedie writes:
> On Fri, 26 Jun 1998 03:15:30 -0400 (EDT), "Albert D. Cahalan"
> <acahalan@cs.uml.edu> said:

>> Explain how to do real-time IO bandwidth reservations with a
>> filesystem that is unaware of the underlying structure.
>
> If you want to start that game, explain where you got the idea that
> Linux's filesystems and IO subsystems were _ever_ designed to be able to
> make guarantees about IO bandwidth at all!

I didn't get that idea, since Linux is not designed for that.
Linux was not designed to be portable, was not designed for SMP,
was not designed to real-time scheduling, and was not expected
to ever support SCSI.

My concern is that implementation choices made now will mean that
future hackers will have to rip up more code than they might
otherwise need to. (not that it would be easy in any case)

Anyway, how else would it be done? You said that the "simple"
solution (dumb filesystem on top of generic LVM) wasn't going
to work so well, didn't you? Have you changed your mind?

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu