Revision # theorhetical upper limit

[David CM Weber <da...@backbonesecurity.com>]

I'm just wondering what the max_value for revisions are for a
repository.  I'm assuming that it is architecture dependant.  An x86
system would be 2^32 possible revisions.  IA64 would be 2^64 possible
revisions.  It this correct, or is it smaller (bigger).  
 
Just wondering about this for a high traffic repository, and I just need
some ammo to fight politics.
 
Thanks
 
 

Re: Revision # theorhetical upper limit

[Jan Hendrik <ja...@bigfoot.com>]

Concerning Re: Revision # theorhetical upper l
kfogel@collab.net wrote on 19 May 2004, 8:11, at least in part:

>    "Judging developers by the number of changes they make is like
>     judging a legislature by the number of laws it passes."
> 
> Hey, I said it, so I can put quote marks around it. :-)

Had you the socialist-green ruling Germany since 1998 in mind? 
<GG>

Jan Hendrik

---------------------------------------
Freedom quote:

     Yet any time you and I question the schemes of the do-gooders,
     we're denounced as being opposed to their humanitarian goals.
     It seems impossible to legitimately debate their solutions
     with the assumption that all of us share the desire to help
     the less fortunate. They tell us we're always 'against',
     never 'for' anything. 
                -- Ronald Reagan, October 27, 1964


---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@subversion.tigris.org
For additional commands, e-mail: users-help@subversion.tigris.org

Re: Revision # theorhetical upper limit

[kf...@collab.net]

Reid Spencer <re...@x10sys.com> writes:
> Perhaps a more realistic score board would be to publish the projects
> with the highest number of commits per month. Not exactly an accurate
> reflection of project activity since the contest would tend to foster
> lots of smaller, otherwise unnecessary, commits. But, it might be an
> interesting benchmark on SVN itself.  Projects could strive for 10,000
> or 100,000 commits per month.
> 
> Perhaps I'll write a little C program to compute the number of commits
> in a given time range.

Feel free... but I must admit, personally I wouldn't want to encourage
that sort of race

   "Judging developers by the number of changes they make is like
    judging a legislature by the number of laws it passes."

Hey, I said it, so I can put quote marks around it. :-)

-Karl

---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@subversion.tigris.org
For additional commands, e-mail: users-help@subversion.tigris.org

Re: Revision # theorhetical upper limit

["C. Michael Pilato" <cm...@collab.net>]

Reid Spencer <re...@x10sys.com> writes:

> Kinda makes me wonder if there should be a contest to reward the
> first *legitimate* open source project to reach the one million
> commit mark.  One hundred developers doing 10 commits a day could
> get there in 2.7 years.

Actually, we joked in the office today about writing a quick Python
bindings script to just make and commit txns rapidly, just to see what
the error case looks like when we finally run out of revisions.

---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@subversion.tigris.org
For additional commands, e-mail: users-help@subversion.tigris.org

Re: Revision # theorhetical upper limit

[Reid Spencer <re...@x10sys.com>]

On Tue, 2004-05-18 at 12:35, Ben Collins-Sussman wrote:

> So you'd have to commit every second for the next 60 years or so, to use
> up all 2 billion revisions.  On a 64-bit machine, it would take much
> longer.

Hmm, yeah about 585 billion years! :). 

Kinda makes me wonder if there should be a contest to reward the first
*legitimate* open source project to reach the one million commit mark.
One hundred developers doing 10 commits a day could get there in 2.7
years.

For a billion commits, a thousand developers doing 10 commits a day
could get there in 274 years :)

Perhaps a more realistic score board would be to publish the projects
with the highest number of commits per month. Not exactly an accurate
reflection of project activity since the contest would tend to foster
lots of smaller, otherwise unnecessary, commits. But, it might be an
interesting benchmark on SVN itself.  Projects could strive for 10,000
or 100,000 commits per month.

Perhaps I'll write a little C program to compute the number of commits
in a given time range.

Reid

Re: Revision # theorhetical upper limit

[Ben Collins-Sussman <su...@collab.net>]

On Tue, 2004-05-18 at 14:18, David CM Weber wrote:
> I'm just wondering what the max_value for revisions are for a
> repository.  I'm assuming that it is architecture dependant.  An x86
> system would be 2^32 possible revisions.  IA64 would be 2^64 possible
> revisions.  It this correct, or is it smaller (bigger).  

You are correct.  svn_revnum_t is a signed 32-bit integer on most
systems ("long int" in C).

So you'd have to commit every second for the next 60 years or so, to use
up all 2 billion revisions.  On a 64-bit machine, it would take much
longer.





---------------------------------------------------------------------
To unsubscribe, e-mail: users-unsubscribe@subversion.tigris.org
For additional commands, e-mail: users-help@subversion.tigris.org