Red Hat and the other competitors are working on added technology to make their virtual machine technology more desirable than all of the others
The ability for virtual machines to utilize all or a specific number of physical processors or cores so that ever more complex applications can be supported. It is now expected that virtual machines can execute on 16, 32 or even 64 virtual processors, based upon available physical hardware and the needs of a given application.
The ability to assign applications to specific processors to achieve needed levels of performance
The ability to place virtual machines that make up a single complex workload on several different physical systems to improve overall performance and not create performance problems when multiple high resource consuming tasks end up competing for the same physical resources on a single system
The ability for templates to be developed that fit the requirements for specific workloads so that new virtual machines can be spun up and provisioned quickly to address growing workload requirements
The ability for virtual machine key performance indicators to be monitored and managed using tools that use standards such as the simple network management protocol (SNMP) by exposing a standard management information block (MIB). This makes it possible for just about any management tool or framework to work with the virtual machines.
The ability for virtual machines and the data they’re managing to reside in virtualized storage. This means that important data can reside in just about any type of storage, just about anywhere, and that virtual machines can be moved from one physical processor to another. It also means that organizations can choose storage network media and the type of storage used to best fit their budget and needed performance profile.
Making available virtual machine movement software so that executing virtual machines can be moved from one physical machine to another without losing data or crashing. This function can be the foundation for application clusters, high-availability clusters and even sophisticated back up and disaster recovery functions.
Making virtual machines fit nicely (read integration) into a growing number of cloud computing frameworks, such as AWS/Eucalyptus, OpenStack or CloudStack. As organizations increasingly organize their internal resources as on-premise clouds or utilize resources offered by cloud services providers, this level of integration can become a key requirement when organizations purchase virtual machine software.