When the cloud server finished downloading technology and Memory on Demand enabled, the memory is adjusted depending on the amount of memory. However, this technology is tied to an agent running inside the virtual machine, which sends information about the current memory usage of mod-server (outside of the virtual machine). Therefore, to respond to the lack of memory the system can start only after the mod-agent (it starts with the other services in the cloud server load). In some cases, there is a need to start immediately with the "Head Start" that is, from the beginning to allocate a significant amount of memory.
The default value is set in the value static_min that is the minimum amount. Now, this value can be changed explicitly.
There are three strategies to use:
The server runs with a minimum of memory, and then starts to increase the amount of memory used as the data load or the load increases. So behave cloud servers by default (for which setting is not changed MoD).
The server runs with inflated the value of the memory, and then quickly "blown away" to the desired server. This configuration prevents the sudden lack of memory request large memory database or close an application that allocates large chunks rather than gradually.
The volume of home storage exhibition about the desired value - in this case mod works fine tuning mode, that is, increases / decreases as circumstances change.
If the technology of Memory on Demand is disabled, this value allows you to manually change the current value of the allocated memory (ranging from static_max to static_min) without rebooting, which can be useful in terms of technologies that do not like change memory (Java, for example).
