threads and VMS [Re: dce/rpc "client" api]

[Luke Kenneth Casson Leighton]

> > In this case the completion of the accept() call would 
> > cause a new thread, not a process to start up.  Much 
> > less overhead, so the scalability is better.
> 
> John,  I'm not sure I foolow you here, but I do know
> that the issue of a thread smbd server has been hashed around
> quite a lot in the past.

there may actually be a technical / speed issue that would, with some
performance penalty, speed samba up, for small numbers of clients and/or
for workstations that are running large numbers of simultaneous
programs/threads, where each program/thread on said workstation is making
significant usage of a single samba server.

[i.e. it's a multiplexing issue].

however, this cannot be proven without access to the [multi-threaded]
benchmark tool that demonstrates that samba has a serious
number-of-transactions problem [but an overall data transfer rate of appx
20-50% better than NT :)]

lukes