Re: [boost] [sort] Re: [review] Formal review period for Sort library begins today, November 10, and ends Wednesday, November 19

Subject: Re: [boost] [sort] Re: [review] Formal review period for Sort library begins today, November 10, and ends Wednesday, November 19
From: Vladimir Prus (vladimir_at_[hidden])
Date: 2014-11-10 03:34:23

On 11/10/2014 10:54 AM, Mathias Gaunard wrote:
> On 10/11/2014 08:35, Vladimir Prus wrote:
>
>> I am not sure the library is useful. For very few users, sorting 100kb
>> arrays of integers is a performance
>> bottleneck. And if it's truly a performance bottleneck, one would
>> probably ponder cache architecture, or use of GPU, or
>> a zillion other tricks. Like this paper from 2010 appears to do, for
>> example:
>>
>>      http://dl.acm.org/citation.cfm?doid=1807167.1807207
>
> I have personally seen cases where sorting speed is a bottleneck.
>
> However, there are implementations that are still quite faster than this library (I have one that is at least 3.5 faster than std::sort),
> and that fares better with small sizes.
>
> Nevertheless, it would be interesting to have a radix sorting library implemented in generic C++.

I suppose Boost implementation of radix sort would be fine, as that's fairly classic algorithm. What I'm concerned about this
prosed library is that is not exactly classic and not quite bleeding edge either.

-- 
Vladimir Prus
CodeSourcery / Mentor Embedded
http://vladimirprus.com