Design question - Sharing of single object by multiple processes

[Mike D]

Steve,

Thanks for the response. My question really comes down to, as you suggested,
premature optimization.

It is more for my own understanding than a current practical use.

If an object is loaded into memory and other threads(or processes) can
recieve a pointer to this location, would this not be more efficient than to
load a new one for every unique request? Wouldn't a method such as this
prevent bottle necks in a read heavy environment?

Perhaps some kind of caching implementation is more along the lines I'm
thinking on?


> I want to have an object in memory. This object will be loaded on
> application start up and will contain a list of objects. These will be
> obtained from (most likely) the file system.
>
>>So "application startup" is different from "process startup"? Your
>>application is a group of co-operating processes? Really?

I think I should have said process start up, for example a long running
process that can supply the object to new ones, loaded on webserver start
up.

I appreciate comments, It is purely so I can understand python and program
design better.

Regards,

Mike



On Feb 1, 2008 12:09 AM, Steve Holden <steve at holdenweb.com> wrote:

> Mike D wrote:
> > Hello, I've just picked up the Python language and am really enjoying
> it.
> >
> > I've just signed up to this mailing list and I'm looking forward to
> > taking part in some discussions.
> >
> > My first post is a question I've been pondering for the last couple of
> days:
> >
> > For relatively static data (such as a list), is it possible to load a
> > single instance and have many python processes access it?
> >
> First there's the problem of having multiple processes access any kind
> of shared resource. So your question makes me wonder whether you mean
> processes, or threads. Are you *sure* you mean processes?
>
> > I want to have an object in memory. This object will be loaded on
> > application start up and will contain a list of objects. These will be
> > obtained from (most likely) the file system.
> >
> So "application startup" is different from "process startup"? Your
> application is a group of co-operating processes? Really?
>
> > My reasoning is: I want to prevent a file system or database fetch each
> > time as it seems unnecessary.
> >
> It might also seem unnecessary to start the Python interpreter for each
> process, but you are going to have to ...
>
> > Is this possible? In Java I'm pretty sure this could implemented with an
> > object factory and static methods. My understanding is that python
> > invokes a separate process for each request however.
> >
> Your understanding is clearly informed by some information you have
> omitted to share with us.
>
> > If possible, what would be a good way of providing a new structure as it
> > is updated.
> >
> > If anyone could give me some advice or point me in the correct direction
> > I'd really appreciate it.
> >
> > Thanks in advance,
> >
> More answers, please. There are mechanisms such as Pyro (Python remote
> objects) that allow inter-process method calls, but I am far from
> convinced that's what you really want (or need, anyway). Perhaps yo
> could let us know a little more about what it really is you are trying
> to do, and convince me, at least, that this isn't just a case of
> premature optimization.
>
> regards
>  Steve
> --
> Steve Holden        +1 571 484 6266   +1 800 494 3119
> Holden Web LLC              http://www.holdenweb.com/
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-list/attachments/20080201/676481ae/attachment.html>