I'm using Primocache primarily to cache my game disk, hdd, to a small 80 gig ssd drive, using L2 read cache only. I've noticed a very nice boost in loading performance for the games I play, so I can say it works perfect for my needs.
My question is this - I have one game that has almost all of its data stored in a big .dat file. The file in question is roughly 42 gigs in size, and is actually a big special archive containing several thousands of files.
What I'm wondering is this - is it possible for primocache to cache parts of that file, or will it be skipped completly if it can't fit? I know Primocache uses block caching, but I'm not really sure what this actually means.
Also, when moving on to another game, how does Primocache decide on what data to keep in the cache, and what data to purge? How long does it take for the program to start purging 'obsolete' data in the cache, to make room for new data? In short, how much goes into learning user behavior, so that the program won't simply purge an existing cache in favor of a "new" one if you for example run a new game a single time, then decide to go back to the "old" games again.
Thank you very much for this program!
Caching of big files Topic is solved
Re: Caching of big files
Primocache will cache parts of a large file yes. It works on a "block" principle, which is a multiple of the cluster size on the drive.
For example, on a 4k cluster size volume, you can set Primocache to use a 4k block size (1:1), or an 8k block size (2:1), etc. Primocache keeps the blocks it sees as being the most accessed, which in the case of large files will be smaller portions of it. A larger 10GB file broken up into 4k blocks, would mean many many blocks for Primocache to work with, and it would keep those it saw the most activity with.
From that perspective it isn't file caching software, it's volume caching software working with blocks of data, which are some multiple of cluster size.
For example, on a 4k cluster size volume, you can set Primocache to use a 4k block size (1:1), or an 8k block size (2:1), etc. Primocache keeps the blocks it sees as being the most accessed, which in the case of large files will be smaller portions of it. A larger 10GB file broken up into 4k blocks, would mean many many blocks for Primocache to work with, and it would keep those it saw the most activity with.
From that perspective it isn't file caching software, it's volume caching software working with blocks of data, which are some multiple of cluster size.