RE: Fundamental architecture of a proxy implementing keep-alive.

From: Dancer [mailto:dancer@zeor.simegen.com]
> > > I mislike me the use of threads. Having written a couple small
> > > http proxies recently, I find that state-machines seem to work
> > > better than threaded models.

"James A. Donald" wrote:
> > Are you suggesting thread and one state machine for each
> > client-proxy connection, or one thread and one state machine for
> > all client-proxy connections?

From: Dancer [mailto:dancer@zeor.simegen.com]
> A single thread of execution for everyone. select() or poll()
> coupled with non-blocking connect()s and such.

I find it odd that the non blocking connect is "less hungry", since if
we are polling then the thread runs continuously, and CPU usage should
remain stuck permanently at close to 100%, whereas with blocking
connects, and a separate thread for each connection, CPU usage should
drop to zero when there is nothing to do.

My proxy is not plannned to run on machine dedicated solely and
exclusively to proxying.
Received on Thu Aug 26 1999 - 11:30:01 MDT