>I'm very interested in the viability of large Linux servers, and I'd
>love to hear what you think is missing. My pet items are support for
>very large (> 10K) fds per process, and I/O completion ports (or
>something like that; sct's SIGRTIO system would work fine for me).
I'd like to see enough statistics out of the low level disk drivers to
get a usable iostat. The information from iostat is vital, IMO, for
managing a large server and being (roughly) able to predict when you are
going to be over capacity before it happens. Linux's total reads and
write's per disk just don't cut it. We need to know service time for
writes, %busy, total reads and writes per second and amount read and
written per second at the very least.
--
Artificial Intelligence stands no chance against Natural Stupidity.
GAT d- -p+(--) c++++ l++ u++ t- m--- W--- !v
b+++ e* s-/+ n-(?) h++ f+g+ w+++ y*
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu