Re: scalable web gallery

At 8:51 AM +0800 7/12/08, paragasu wrote:

Why not use a hash table based upon the name of the file and distribute files that way? sorry tedd, i don't really get it. but i am interested to know more about the hash table concept. can you explain a little bit more..

paragasu: A hash table is simply a result from an algorithm where you take the name of the file and produce a number from it. It can be any computation you want that produces a number -- such as just totalling the number of letters in the name. For example, tedd.gif would compute to 4, whereas Rob.gif would be a meager 3, and Daniel.gif would be a six (which is probably over-rated for him anyway). :-) Now, just use our respective number for storage in directories -- tedd.gif would be stored in directory 4, Ron.gif would be stored in directory 3, and Daniel.gif in directory 6. If you store the names of the files in a database, then you don't need to store the paths as well because the paths can be computed from the file names -- do you see? Also, you don't need to create a directory until there's actually a need for one. This is old logic for creating memory structures where we weren't forced to allocate memory beforehand -- it's like a large array with holes in it. It's kind of slick when you can apply it to other stuff. Now, the example I gave was pretty simple. However, a good hash algorithm is one that distributes things evenly, and that's a bit more complicated -- I'll leave that for another time. <now returning you to your normal television viewing pleasure> :-) Cheers, tedd -- ------- http://sperling.com http://ancientstones.com http://earthstones.com