Memory management on current operating systems can be a bit tricky to explain; the main thing to understand for this tip is that the memory assigned to an aplication can be either Physical Memory (RAM) or it can be disk-based (memory that has been swapped back to disk). It is usually up to the OS to determine which of the two types of memory allocated and the OS can further swap data between these two types of memory as it sees fit.

When you combine the amount of RAM used plus the amount of memory swapped to disk you get what we refer to as Used Virtual Memory; this is the overall total amount of memory used by the 4D Application.

32 bit applications (such as 4D Server 32 bit, 4D Remote, 4D Singleuser, or 4D Volume Desktop) can only address 2^32 bytes of memory (4GB); if a 32 bit application attempts to address memory beyond this limit the application WILL crash.

The limit of 2^32 bytes of memory is not related to RAM nor SWAP individually, however the limit is related to the accumulated value of both RAM+SWAP combined. Because of this, when investigating something like a memory leak, the Used Virtual Memory statistic is more important then the amount of Used Physical Memory.