Specifications

Best Practices for Virtualizing and Managing Exchange 2013
15
15
NUMA is a memory design architecture that delivers significant advantages over the single system bus
architecture and provides a scalable solution to memory access problems. In a NUMA-supported
operating system, CPUs are arranged in smaller systems called nodes (Figure 4). Each node has its own
processors and memory, and is connected to the larger system through a cache-coherent interconnect
bus.
14
Figure 4: NUMA node (processor and memory grouped together)
Multiple NUMA nodes can exist in a host system (Figure 5). In the context of multiple nodes:
 Local memory is attached directly to the processor (grouped into a node).
 Remote memory is local to another processor in the system (another node).
This grouping into nodes reduces the time required by a processor to access memory (locally located), as
the processor can access local memory faster than remote memory.
15
Figure 5: Multiple NUMA nodes on a single host
Root/Host Reserve
Root reserve or host reserve is the amount of memory that is reserved for the root partition and is
guaranteed to be available to the root partition. It is not allocated to any of the virtual machines running
in the child partition. Hyper-V automatically calculates root reserve based on the physical memory
available on the host system and system architecture.
16
Best Practices and Recommendations
The root partition must have sufficient memory to provide services such as I/O virtualization,
virtual machine snapshot, and management to support the child partitions. Hyper-V calculates an
amount of memory (known as the root reserve), which is guaranteed to be available to the root
partition. This memory is never assigned to virtual machines. Root reserve is calculated
automatically, based on the host’s physical memory and system architecture.