Date: Sat, 13 Mar 2004 22:15:10 -0600
From: John Goerzen <jgoerzen@complete.org>
To: gopher@complete.org
Subject: [gopher] Re: Gopherd or pygopherd?
Message-ID: <20040314041510.GA1361@complete.org>
References: <Pine.LNX.4.58.0403131538400.3779@tanatos>
 <20040313210311.GA10755@complete.org>
 <20040313181126.2e064e3c.markov@monmouth.com>
 <20040314031639.GB30332@complete.org>
Mime-Version: 1.0
Content-type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20040314031639.GB30332@complete.org>
User-Agent: Mutt/1.5.5.1+cvs20040105i
Content-Transfer-Encoding: 8bit
Sender: gopher-bounce@complete.org
Errors-to: gopher-bounce@complete.org
Precedence: bulk
Reply-to: gopher@complete.org

On Sat, Mar 13, 2004 at 09:16:39PM -0600, John Goerzen wrote:
> I don't think PyGopherd is a resource hog.  It will read all your UMN

I should add some info.  On the quux.org server, the pygopherd process
consumes 3.5MB of RAM.  148K of that is shared (meaning it is not
actually extra RAM that is used).  On that server, which does dip into
swap, all but 150K of the remainder tends to remain paged out.

This is with most all of the modules loaded and Python optimizations
turned off.  If you were judicious about the loaded modules, and used
Python optimizations, your results will probably be even better.

Should be no problem on a 32MB box.

PyGopherd can work with either forking or threading.  You may want to
experiment to see which method is kinder to your resources.

-- John