[Date Index] [Thread Index] [Author Index]
Re: memory question
On 8/9/08 at 7:45 AM, jasch at crustal.ucsb.edu (Jan Schmedes) wrote: >I am running Mathematica 6.0.3 on a Mac Pro with 4 Gig memory. I >have fortran binary files that I can read in correctly (numbers of >order 10^6) using Import[...,"Real32"]. But for larger problems, the >size of the binary file is about 80 Mb, all memory is used (if i hit >top, VSIZE is 17G) and only 2% of the CPU are used, so it basically >can not handle the file. Is there a way to decrease the precision at >which the file is read in, for my purpose the precision does not >have to be very high? Even if there is a simple way to decrease the precision as you read the data in this will not solve your problem. I am sure you will find minimum storage of floating point numbers occurs when you use machine precision which on your machine I believe will be 64 bits. Using Mathematica's arbitrary precision technology will increase storage requirements since there is additional information that needs to be stored. The only real solution to very large problems is to read part of the data in, process that part and repeat until you've processed all of the data. In this way, the limit becomes hard drive storage rather than RAM. Of course the total time to process the data will increase due to overhead associated with managing portions of the data and I/O speed of the hard drive. But this might be all that bad as a custom routine to read in the data might perform more efficiently than Import.