File-system caches are a mechanism used by most of today's operating systems to avoid redundant disk access. In order to accomplish this, a part of the computer's physical memory (RAM) is used to buffer data that is read from and written to disk. This memory region is then called a "cache".

Now, if an application accesses disk data that is still residing in the file-system cache, the operating system does not have to read it from the disk, but can simply take it out of the cache, which is a lot faster.

On the other hand side, if an application is writing data to disk, OS/2 can copy this data to the cache first and delay writing it to disk until the disk is ready for accepting it. This is called "lazy writing".

Do not confuse the file-system caches, which we are dealing with here, with the other caches that exist on your system, such as the processor caches and maybe other caches that are built into your hard disks. These are part of the computer's hardware and not described here.

OS/2 implements different caches for each file system that is installed. This means that all FAT drives will use one cache, and all HPFS drives a different one. These caches are separately configured with different CONFIG.SYS settings, which you may alter on the "HPFS" and "FAT" pages in the "OS/2 Kernel" object.

Press the "Help" button there to receive a more detailed description of the different settings.