Pete Templin <[EMAIL PROTECTED]> writes:
> What kind of performance do you get with your md devices?
Haven't really paid close attention.
> It seems to me (operating in a vacuum not knowing what other disks
> you have) that you'd do better by shifting your hdb (primary slave
> IDE disk) to hdc (
On 15 May 1997, Rob Browning wrote:
> Pete Templin <[EMAIL PROTECTED]> writes:
>
> > As my system is currently laid out, I'm not suffering from any shortage of
> > swap _space_ (240MB allocated, max seen in use is 33MB), I'm merely trying
> > to sneak out the best possible performance without sp
On 15 May 1997, Rob Browning wrote:
> Pete Templin <[EMAIL PROTECTED]> writes:
>
> > As my system is currently laid out, I'm not suffering from any shortage of
> > swap _space_ (240MB allocated, max seen in use is 33MB), I'm merely trying
> > to sneak out the best possible performance without spe
Pete Templin <[EMAIL PROTECTED]> writes:
> As my system is currently laid out, I'm not suffering from any shortage of
> swap _space_ (240MB allocated, max seen in use is 33MB), I'm merely trying
> to sneak out the best possible performance without spending a buck.
In the end I imagine you'll be m
On 15 May 1997, Miquel van Smoorenburg wrote:
> You can stripe swap without using md - the kernel does it for you
> ever since Linux 1.3.2 :)
>
> Read the "swapon" manpage, option "-p".
The kernel may allocate pages on a round-robin basis, but I'd like to tuck
my swapspace into my md disk for b
In article <[EMAIL PROTECTED]>,
Daniel Quinlan <[EMAIL PROTECTED]> wrote:
>Pete Templin <[EMAIL PROTECTED]> writes:
>
>> I'm having some difficulty creating "special" swapfiles to be able to test
>> swapping to/from an "md" device (raid0). I've followed the manpage hints
>
>You must give the full
6 matches
Mail list logo
