Host Details - KVM Test

This test provides the administrators with a series of configuration measures such as the number of sockets, cores, nodes and threads. Additionally, this test focuses on the number of virtual CPUs and the CPUs that are active on the KVM server host. This way, this test enables the administrators to judge their KVM server infrastructure and load the virtual machines on the server accordingly!

Target of the test : A KVM server

Agent deploying the test : An internal agent

Outputs of the test : One set of results for the KVM server that is to be monitored.

Configurable parameters for the test
Parameter	Description
Test Period	How often should the test be executed.
Host	The IP address of the host for which this test is to be configured.

Measurements made by the test
Measurement	Description	Measurement Unit	Interpretation
CPU sockets	Indicates the total number of CPU sockets that are available on this server.	Number	A high value for this measure indicates that the administrators may still load additional virtual machines on the KVM server.
Cores per socket	Indicates the total number of cores available in each socket of this server.	Number
Active CPUs	Indicates the number of CPUs that are currently active on this server.	Number	A high value is desired for this measure.
Virtual CPUs	Indicates the number of virtual CPUs that are available on this server.	Number	A high value for this measure indicates that the virtual machines are adequately allocated with memory resources.
CPU frequency	Indicates the average frequency at which the CPUs are operating on this server.	Mhz
Numa cells	Indicates the number of nodes i.e., NUMA cells available on this server.	Number	Non-uniform memory access (NUMA) is a memory architecture implemented in multi-processor systems where the memory is spread among processors instead of being central like in standard Symmetric Processing Servers (SMP). This design reduces the risk of bottleneck on the links between CPUs and memory. A guest virtual machine on a NUMA system can be pinned to a processing core so that its memory allocations are always local to the node it is running on. This avoids cross-node memory transports which have less bandwidth and can significantly degrade performance. A high value is desired for this measure.
Threads per core	Indicates the average number of threads that are available per core on this server.	Number