Re: [DotGNU]DotGNU task list

[Bill Lance]

--- Jonathan P Springer
<address@hidden> wrote:
> 
> I don't want to lose sight of the "address@hidden"
> distributed computing
> possibilities of the architecture.  Since much of
> what DotGNU envisions
> requires P2P-like distributed networking,
> distributed processing
> services are a distinct possibility.  For example:
> 
> 1.  I want to be rich.
> 
> 2.  I devise some sort of Monte Carlo derivative
> modelling scheme and
> code it in some sort of VRS/DEE aware
> implementation.
> 
> 3.  Part of that implementation is the function 
> "myReturnType ReallyComplexAlgorithm(MyInput1 i1,
> MyInput2 i2)".
> 
> 4.  I call some (hypothetical) service
> "distributedComputation(ReallyComplexAlgorithm,
> distributedInputs[])".
> 
> 5.  The distributed computation starts doling the
> computations (RPCXML's
> containing bytecode and input?) out through a
> network of P2P-connected 
> VRS/DEEs.  The algorithm for distribution is left as
> an exercise for 
> the student :-).
> 


Now this should be possible in a VRS.  You may have to
organize and tune a Cluster to serve this task
yourself, though.  Let me explain.  The VRS is a set
of technologies and methods that allow people for form
p2p clusters tuned and adapted to meet a variety of
purposes.  It is NOT a single, ubiquitous open boundry
system like Gnutilla.  Each VRS Cluster is assembled
and tuned to meet a specific set of objectives.  Doing
this is an administrative job, not a technical one.  

The VRS homepage mentions just a few of the possible
configurations.  As we continue to work on the VRS,
other possibilities arise.  Your example points to
several of these.

1) A commercial provider contract with node host
owners for VRS resources (CPU time, disk space and
bandwidth) to form a Cluster.  Then you contract with
them to install and expose your component calls. 
These can be configured so only you can use them.  Or,
you may wish to expose them to your clients or the
general public.

2) You yourself, could organize such a Cluster amoung
you clients directly.  You could charge fees to
whatever degree your customers felt you add true
value.

3) You could form your own private Cluster of machines
salvaged from some luckless business converting to XP.
Then run your components accross the Cluster.

If the VRS design works out, you should be able to use
the same Datasets in all of the above scenarios, and
be free to move from one to the other at will.


__________________________________________________
Do You Yahoo!?
Send FREE Valentine eCards with Yahoo! Greetings!
http://greetings.yahoo.com