All,
FYI. Earlier versions of Windows NT had this same problem (and it still
does). The way they got around it was to "cheat" by cranking up the
priority of the console an Windows GUI subsystem anytime someone hit a
key or moved the mouse. In fact, an NT server is still heavily loaded,
but by making the display look and feel "snappy", it presents the
illusion that the server has great responsiveness under heavy load.
This stuff in NT is all hard coded (Since their GUI is tightly
integrated with the OS), and if Linux wanted to do the same, the X
Server and bash would probably be the places to do this .....
Jeff
Andrea Arcangeli wrote:
>
> On Wed, 8 Mar 2000, Nicolas MONNET wrote:
>
> >I've noticed that under heavy loads, and actually, not so heavy loads,
> >Linux's responsiveness is effectively very poor.
>
> Which kernel version are you using?
>
> >[..] The
> >system isn't even swapping, [..]
>
> Supposing you are using 2.2.x if it wasn't VM related it's probably the
> elevator starvation thing. Could you try again with 2.2.14aa10 (it
> includes elevator-starvation-6.gz):
>
> ftp://ftp.*.kernel.org/pub/linux/kernel/people/andrea/kernels/v2.2/2.2.14aa10.gz
>
> For reference the elevator-starvation-6.gz patch alone is here:
>
> ftp://ftp.*.kernel.org/pub/linux/kernel/people/andrea/kernels/v2.2/2.2.14aa10/elevator-starvation-6.gz
>
> >It's not like there's a bunch of runaway processe: at most, 5 processes
> >are running.
>
> To be sure about what's going on could you provide me a `vmstat 1` log
> of before/during/after the performance drop?
>
> Andrea
>
> -
> 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/
-
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/
This archive was generated by hypermail 2b29 : Wed Mar 15 2000 - 21:00:14 EST