[SciPy-dev] [Numpy-discussion] Memory mapped files in scipy core

[Perry Greenfield]

> Each of these things implicitly changes the layout of the underlying 
> file.   Whether or not these features get used or justify the complexity 
> is another matter.   Because of 32-bit address space exhaustion and swap 
> file  issues, memory mapping was a disappointment at STSCI so I'm 
> doubtful we used these features ourselves.
> 
> Regards,
> Todd
>
Let me add that I believe this situation of memory mapping being not
terribly useful now is (or should be) temporary. Basically what has
happened is that the size of file that is interesting to memory map
is larger than the current 2GB limit. When this limit is removed, I
believe that there will be renewed interest in using memory mapping
with data sets that are much larger than physical memory. We already
occasionally encounter such data sets now but can't use this as a 
means of accessing the data.

Perry