Re: [boost] [serialization] fast array serialization (10x speedup)

From: David Abrahams (dave_at_[hidden])
Date: 2005-11-20 19:01:42

bwood <brass_at_[hidden]> writes:

> Matthias Troyer wrote:
>>> 
>>> i) somehow I doubt that this archive type has been tested with all
>>> the serialization test suite. Instructions for doing so are in the
>>> documenation and the serialization/test directory includes batch
>>> files for doing this with one's own archives. Was this done? What
>>> where the results? With which compiler? It costs nothing to do
>>> this.
>>
>>
>>Just ask if you had a doubt. The short answer is "I have done this". 
>>After adding the fast array serialization to the binary and 
>>polymorphic archives, I ran all your regression tests, without any 
>>problem (using gcc 4 under MacOS X). 
>
> Have you done this with vector also?  I would be surprised
> if you have.  

He has.

> It seems to me there is a standards issue to 
> be addressed in this area.  It is no problem to send the
> data in a vector of int as a unit, but I don't know of a 
> way to add/receive the data into a vector correctly.  It is 
> possible to stuff the data into the correct place, but the
> vector's internal state/size does not get set properly since 
> you didn't call push_back or some other method to get the 
> data into place.  Adding something like push_n_back to vector 
> might help when it comes this.  

You resize first and then copy the data in.  It's a factor of 2 slower
than the ideal, but it's still way faster than doing what the library
does today.

-- 
Dave Abrahams
Boost Consulting
www.boost-consulting.com