The number one capacity constraint in most virtualized environments is memory. If memory isn’t managed properly, an environment’s performance and possibly the data center’s ROI will suffer.
Unfortunately, some virtualization administrators are unfamiliar with which memory metrics should be used, as well as how to use them to correctly size virtual machines and troubleshoot performance problems..
This paper explains virtualized environment memory management and recommends memory metrics to use for virtual machine (VM) memory allocation and performance issue monitoring. It also discusses how to make these assessments when the “large memory pages” setting is active, which makes memory management even more complicated.