[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Monotone-devel] [ANNOUNCE] monotone 0.19
From: |
Nathaniel Smith |
Subject: |
Re: [Monotone-devel] [ANNOUNCE] monotone 0.19 |
Date: |
Thu, 5 May 2005 04:15:05 -0700 |
User-agent: |
Mutt/1.5.9i |
On Thu, May 05, 2005 at 11:29:43AM +0200, Nico -telmich- Schottelius wrote:
> Hello Nathaniel, list,
>
> the results look very promising, still there's one value I am
> a bit concered about:
>
> Nathaniel Smith [Tue, May 03, 2005 at 04:36:56AM -0700]:
> > - time to do a fresh pull of collection net.venge.monotone with 0.19
> > (despite the collection being >25% larger than when 0.18 was
> > released): 15m30s
>
> Do you know why it takes so long? Is it because too much cpu load
> or because of the masses of data transferred?
>
> If I assume 1Mbit/s, it makes 125KiB/s. And if I do now calculate
> ((15*60+30) * 125)/1024 with bc, it tells me I download about
> 113MiB. Is that true?
Well, that number is actually CPU load directly, since when putting
together the notes my machine had 2-3 builds running in the
background, so more proper benchmarking wasn't really possible :-).
However, the initial pull is on the whole mostly CPU bound; monotone
is rather busy doing hashing, compression, uncompression, xdelta
calculation and application, and extensive validity checking on all
the new data.
-- Nathaniel
--
"...All of this suggests that if we wished to find a modern-day model
for British and American speech of the late eighteenth century, we could
probably do no better than Yosemite Sam."