Users are reporting CPU related performance problems on the virtual machine Email-Prod throughout the day.
The resource settings for the VM and the ESXi host the VM is running on are shown below:
VM:
Email-Prod
vCPUs: 4
Memory: 96GB
ESXi host:
CPU: 2 x 8 Core Processors
Memory: 128GB
NUMA architecturE. 2 Nodes
Which two options would alleviate the observed performance problem for Email-Prod? (Choose two.)
A.
Enable the advanced parameter Numa.PageMigEnable.
B.
Enable the advanced parameter Numa.AutoMemAffinity.
C.
Enable CPU affinity to separate 2 vCPUs for each NUMA node.
D.
Enable CPU affinity to bind all vCPUs to one NUMA node.
Explanation:
Explanation/Reference:
B and C are correct.
http://pubs.vmware.com/vsphere-60/index.jsp#com.vmware.vsphere.resmgmt.doc/GUID-17B629DE-75DF-4C23-B831-08107007FBB9.html
This page actually has the options listed in the answers
https://pubs.vmware.com/vsphere-55/index.jsp?topic=%2Fcom.vmware.vsphere.resmgmt.doc%2FGUID-A420C263-D5B9-4BCA-8015-DE9804B36D21.html
I also looked up this to get a handle on NUMA
https://en.wikipedia.org/wiki/Non-uniform_memory_access
Had the VM had 64GB RAM or less, the preferred design would be ‘D’. But because it has 96GB, the memory spans both NUMA nodes, thus ‘C’.