VSwap

New RHEL6-based OpenVZ kernel has a new memory management model, which supersedes User beancounters. It is called VSwap.

Now you can set two primary parameters: physpages and swappages, while all the other beancounters become secondary.

• physpages

This parameter limits the physical memory (RAM) available to processes inside a container.

The barrier is ignored, and the limit sets the limit.

Currently the user memory and the page cache are accounted into physpages.

• swappages

This parameter limits the amount of swap space which can be used for processes inside a container.

The barrier is ignored, and the limit sets the limit.

The sum of physpages.limit and swappages.limit limits the maximum amount of allocated memory which can be used by a container. When physpages limit is reached, memory pages belonging to the container are pushed out to so called virtual swap (vswap). The difference between normal swap and vswap is that with vswap no actual disk I/O usually occurs. Instead, a container is artificially slowed down, to emulate the effect of the real swapping. Actual swap out occurs only if there is a global memory shortage on the system.

Secondary parameters

Even though these settings are classed as secondary, they are still important settings!

In addition to physpages and swappages being set, it is currently recommended to:

• Set kmemsize to no more than 1/2 of physpages
• Set dcachesize to no more than 1/2 of kmemsize
• Set lockedpages to no more than 1/2 of physpages

VSwap CT config samples (like ve-vswap-256m.conf-sample that come with vzctl are currently implementing these recommendations.

Note: VSwap feature is still being worked on, so these recommendation might change in the future.

Note: As of 042stab042 and later kernels, you are not required to set kmemsize, dcachesize and lockedpages resources, so you can set those to unlimited. For more information, see Kir's blog post on the subject [1].