On Sep 13, 2007, at 2:59 PM, Patrick K. O'Brien wrote:

> Thanks a ton for satisfying my curiosity.  It seems a shame that
> packing is so darn slow, since it brings such obvious benefits to
> the startup time.  So now I'm wondering if there is any room for
> optimization of the packing process.  Anyone from Durus care to
> comment on that?

There is a little room, but not much.  A pack must write out almost
all object
records, so I think it is limited by io speed.

Did you know, though, that a client can connect to a Durus Storage
Server as it is
running and initiate a pack.  The Storage Server continues to serve
requests, and
switches to the packed file when the pack is complete.
Since packing can happen on a live server, I don't see that much
value in trying to
make it faster.