Re: Benchmark results on mdds::multi_type_vector

Hi Kohei,

On 13/12/2019 03:01, Kohei Yoshida wrote:
> I just finished my benchmark testing on mdds::multi_type_vector, and
> summarized my results in this blog post:
>
> http://kohei.us/2019/12/12/benchmark-results-on-mdds-multi_type_vector/

        Wow - that's wonderful work =) I guess any theoretical difference
between binary & interpolation search is moot there too - given how
close it is to the hinted version.

> Hopefully my findings and intepretations make sense.  In short, the
> numbers look great.  The overhead of block shifting is a concern, but
> I'm optimistic that this is going to be a non-issue for the most part.

        Really encouraging; of course - in theory 6.4 is frozen & all that, but
- I'd love to get this into master ASAP and see what the crash tests /
Calc regression tests say (if anything). AFAIR you have amazing ~100%
unit-test coverage of each branch in mdds.

        Given the large wins, what would you think of the idea of pushing to
get this released early ? ;-)

        Thanks !

                Michael.

-- 
michael.meeks@collabora.com <><, GM Collabora Productivity
Hangout: mejmeeks@gmail.com, Skype: mmeeks
(M) +44 7795 666 147 - timezone usually UK / Europe