Release of the Collaboration Package Version 1.0.1a

Release of the Collaboration Package Version 1.0.1a

[Michael Erdmann]

Dear all,

Collaboration 1.0.1a is a simplistic Ada framework which allows to speed up 
computing task by applying a simple parallelism scheme.
  
The framework is not intended to solve all problems of parallelism but
it can be applied to all situations where it is possible to break down 
the result set into disjunct partitions which can be calculated 
independently of other partitions. Anyway it is rather easy to use :-)

You may download the release via:

https://github.com/merdmann/Collaboration/releases/tag/1.0.1a


The software is in a rather early stage; so it lacks good documentation,
error handing and a system wide installation procedure.

Regards
   M.Erdmann

Re: Release of the Collaboration Package Version 1.0.1a

[Randy Brukardt]

How does this differ from Brad Moore's "Parafin" libraries?

                                 Randy.

"Michael Erdmann" <michael.erdmann@snafu.de> wrote in message 
news:52112204$0$14294$afc38c87@news7.united-newsserver.de...
> Dear all,
>
> Collaboration 1.0.1a is a simplistic Ada framework which allows to speed 
> up
> computing task by applying a simple parallelism scheme.
>
> The framework is not intended to solve all problems of parallelism but
> it can be applied to all situations where it is possible to break down
> the result set into disjunct partitions which can be calculated
> independently of other partitions. Anyway it is rather easy to use :-)
>
> You may download the release via:
>
> https://github.com/merdmann/Collaboration/releases/tag/1.0.1a
>
>
> The software is in a rather early stage; so it lacks good documentation,
> error handing and a system wide installation procedure.
>
> Regards
>   M.Erdmann 

Re: Release of the Collaboration Package Version 1.0.1a

[Michael Erdmann]

On Mon, 19 Aug 2013 17:00:06 -0500, Randy Brukardt wrote:

> How does this differ from Brad Moore's "Parafin" libraries?

Parafin tries to provide a broad range of patterns for parallelism.
This is i guess the biggest difference; i am _not_ intending this.
I have just refactored the existing code i am using for a
long time to make it reusable for other people. 

I have done some performance measurements; the difference in 
performance is poorly noticeable. 


/Michael