[distcc] Benchmarking the server-side cache. How bad is 4ms latency to distcc server?

Dan Kegel dank at kegel.com
Mon Feb 6 02:09:29 GMT 2006

On 2/5/06, Martin Pool <mbp at sourcefrog.net> wrote:
> > ... look at a protocol change for large files to try sending
> > just the hash of the source first, and sending the full
> > source only if the server replies with 'sorry, not in the cache'.
> I think that would be a good idea.  It'd also give you the option of
> polling every server to see if any one of them has it.

Polling every server would be too expensive in practice,
I bet.  Probably better to use the hash of the source to
pick a server -- that way we know which one has it.

> > But I thought I'd ask: just how awful is it to have a latency
> > that alternates betwen 0.4 and 4.0 ms, with an average of 1ms?
> > My guess is that it causes a significant overall penalty
> > compared to a uniform 0.4ms latency, but I haven't checked.
> It's going to depend on whether the slow jobs are on the critical path
> for the overall build, which in turn will be influenced by whether
> you're using Make or some other tool better at scheduling.

We're using a nonhierarchical Makefile.   It seems ok scheduling-wise,
but I haven't really looked.

First experiments with lzo seem to show a slight performance loss
when tested without caching, so I guess it's on to the protocol change.
I was hoping to avoid that...
- Dan

Wine for Windows ISVs: http://kegel.com/wine/isv

More information about the distcc mailing list