Boost logo

Boost :

Subject: Re: [boost] ASIO into the standard
From: Ion Gaztañaga (igaztanaga_at_[hidden])
Date: 2014-07-11 15:41:02

El 04/07/2014 12:46, Neil Groves escribió:

>This is evident from the
performance improvements obtained in Boost.Container. I have noticed
though that the improved allocators do not provide superior performance
on Linux systems over the standard implementations.

Hi Neil,

I'm interested in your feedback. Which kind of scenarios are you using
in Linux? The improved allocators should help when inserting ranges or
making copies of complete containers. shrink_to_fit is also much faster
(it does not move objects). Are you using boost::container::allocator or
a more specialized boost::container::adaptive_pool/node_allocator for
node-based container?

You have some basic benchmarks in boost/libs/container/bench if you want
to have some information about the best scenarios.

Linux allocator is based on ptmalloc2 which is based on DLMalloc, so for
usual allocations they should perform similarly. On multithreaded
scenarios the default allocator should be better. In Windows the
performance improvement is noticeable.



Boost list run by bdawes at, gregod at, cpdaniel at, john at