[distcc] DISTCC on Scyld cluster

[Ihar `Philips` Filipau]

On 10/22/07, Marcio Teixeira <mteixeira at numerica.us> wrote:
>
> My concern with this approach is that there is the possibility of
> their being a lot of overhead in migrating hundreds of small "gcc" tasks
> to  the compute nodes one at a time.
>

Uhm... migration is IO bound. Compilation - CPU bound. They should
complement each other nicely.

More than that, migration is local to the system (and most likely
optimized). But sending "hundreds of small gcc tasks" over network already
implies much greater overhead/latencies/etc.

P.S. I personally would try to avoid daisy-chaining because any compilation
trouble would be very hard to diagnose.
-------------- next part --------------
HTML attachment scrubbed and removed