What is your goal for moving away from accessing the 3GB hash? Is it the obvious, of not holding so much memory hostage, or is it something less obvious such as speed performance? Why is it useful to you to hold smaller hashes in memory? Is it the case that you don't really need immediate random access to the entire 3GB hash at any given moment? What does the script actually do?
Would it be reasonable to build an SQLite table so that you're not holding it all in memory at once? SQLite is pretty fast. Or how about a each hash element being represented, instead, by a MongoDB doc?
Dave
In reply to Re: Small Hash a Gateway to Large Hash?
by davido
in thread Small Hash a Gateway to Large Hash?
by lsherwood
| For: | Use: | ||
| & | & | ||
| < | < | ||
| > | > | ||
| [ | [ | ||
| ] | ] |