Re: [Monotone-devel] [ANNOUNCE] monotone 0.19

[Nathaniel Smith]

On Fri, May 06, 2005 at 01:15:29PM +0200, Nico -telmich- Schottelius wrote:
> Nathaniel Smith [Thu, May 05, 2005 at 04:15:05AM -0700]:
> > [initial-push-discussion]
> >  
> > 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.
> 
> Thank you for the information Nathaniel!
> 
> I have just some more questions left:
> 
> - The cpu load/usage is high on
>    a) the monotone server
>    b) the monotone client?

Monotone's basically peer-to-peer.  The only difference between client
in server is who starts the initial handshake, and that the server may
have multiple protocol streams running at once.

The main thing that might make a difference for resource usage is
that servers tend to send out more revisions than they get in, on
balance, and clients the other way around.  Anything that might make a
client slow can also make a server slow, though, and vice-versa.

> - You speak about compression/decompression, are there in general
>   processes, which can be optimized? I don't mean algorithms
>   or something like that.

I'm not sure what you mean by "general process, which can be
optimized...not algorithms" :-).

-- Nathaniel

-- 
"If you can explain how you do something, then you're very very bad at it."
  -- John Hopfield