[Interprocess] Can I use google sparsehash?

Hi, I thought I would find this question already answered on the internet or in previous posts in the mailing list, but my research is fruitless. Hence, since I am not proficient in the way of allocators, I will ask directly: is it possible to use google sparsehash with Boost.Interprocess? According to the info I found sparsehash can be faster than boost::unordered_map, and I could really use the additional speed. Thanks for the help! -- View this message in context: http://boost.2283326.n4.nabble.com/Interprocess-Can-I-use-google-sparsehash-... Sent from the Boost - Users mailing list archive at Nabble.com.

El 31/05/2015 a las 15:06, Malko escribió:
Hi, I thought I would find this question already answered on the internet or in previous posts in the mailing list, but my research is fruitless. Hence, since I am not proficient in the way of allocators, I will ask directly: is it possible to use google sparsehash with Boost.Interprocess? According to the info I found sparsehash can be faster than boost::unordered_map, and I could really use the additional speed.
Thanks for the help!
I think it was never tested before, so my guess is that it won't work. At least iterators can't be placed in shared memory because they store a raw pointer. Best, Ion
participants (2)
-
Ion Gaztañaga
-
Malko