I have matlab version 2009. When i try to read in a file that has 63174528 floats (ie 252698112 bytes), I end up with an array that has lesser number of numbers. I understand that matlab treats numbers as doubles i.e. 8 bytes , but even so, I should have plenty of memory to store 63174528 doubles (i.e. 505396224 bytes)
When i run the memory command in matlab to check out my system resources, here is what i get
Maximum possible array: 28069 MB (2.943e+010 bytes) *
Memory available for all arrays: 28069 MB (2.943e+010 bytes) *
Memory used by MATLAB: 659 MB (6.915e+008 bytes)
Physical Memory (RAM): 16361 MB (1.716e+010 bytes)
* Limited by System Memory (physical + swap file) available.
So as you can see, 28069 MB, is plenty of space to accomodate my file of numbers in an array.
my matlab code is as follows:
Here, I get data as <14598144x1 double> array. I am missing some 600,000 numbers.
Can anyone advise as to what can be done to accomodate 63174528 numbers in an array and why the array is so small when the memory command lists far more memory being available.
PS: the file was written out in the following way in C++:
fwrite((void*)f,sizePerElement,numElements,fp) //dump bytes pointed by a pointer f into the binary file