> > ...I wanted to know how I might go about having judy only store part
> > of the array (or arrays, depending on which is less expensive, memory
> > wise).
> 
> Please elaborate on what you mean by "store part".  The Judy library is
> pretty memory-efficient for the kinds of mappings it supports, but has
> no offline capability.  It's something I studied and wrote about long
> ago, how perhaps we might store and recall an entire array efficiently
> say on a disk drive, but it wasn't trivial (lots of memory fixups would
> be required upon restore) and never implemented beyond the "batch" code
> I wrote.
> 

As in the part that is being worked on in memory is, according to my
estimates a maximum of 300MiB. The rest, some 1800MiB (max) should be
pushed into swap or onto the disk for retrieval in a few seconds,
(the processor should be done with the other section by then). The
method of access is guaranteed as sequential in relation to the
sections to process. A section is not bigger then an array of size
2000*2000*24. Though only about 2000*2000*4 or 5 is expected to be
non-NULL at any one given time.

Thanks again.

------------------------------------------------------------------------------
Dive into the World of Parallel Programming The Go Parallel Website, sponsored
by Intel and developed in partnership with Slashdot Media, is your hub for all
things parallel software development, from weekly thought leadership blogs to
news, videos, case studies, tutorials and more. Take a look and join the 
conversation now. http://goparallel.sourceforge.net/
_______________________________________________
Judy-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/judy-devel

Reply via email to