Re: [boost] [sort] Parallel sorting sub-library mini-review conclusion

Subject: Re: [boost] [sort] Parallel sorting sub-library mini-review conclusion
From: Francisco José Tapia (fjtapia_at_[hidden])
Date: 2016-11-22 03:47:21

Thanks Steven,

I am agree with all the points.

I would like to know the memory used and the best,normal and worst case of
the Asynchronous sorting algorithms.

I must study in deep QuickBook for unify the documentation of the project

Thanks again

Francisco

2016-11-22 3:16 GMT+01:00 Steven Ross <spreadsort_at_[hidden]>:

> The Boost.Sort parallel sorting sub-library mini-review is now complete.
> There was interest expressed in using this library, and all complaints
> about its ability to work were corrected.
> For that reason, I'm going to accept it into the Boost.Sort library,
> subject to these conditions:
>
>    1. Francisco should investigate the reverse-sorted data optimization and
>    see if he can integrate it without significantly impacting performance
> for
>    other data sets.
>    2. Francisco should investigate why Asynchronous quick_intro_sort
>    outperforms Boost parallel sort, and consider adopting the techniques or
>    source of that library call with Christophe's permission.
>    3. We need to integrate the documentation with the rest of Boost.Sort so
>    that it works properly with QuickBook, and the documentation will need
> some
>    editing.
>    4. Francisco will maintain the Parallel sorting sub-library.
>
> Thanks for all the feedback from participants, and Congratulations
> Francisco!
>
> _______________________________________________
> Unsubscribe & other changes: http://lists.boost.org/
> mailman/listinfo.cgi/boost
>