5
VMware
white paper
2. Performance aspects and Best Practice recommendations
This section describes the performance aspects of Fault Tolerance with best practices recommendations to maximize performance.
For operational best practices please refer to the
VMware Fault tolerance recommendations and Considerations on VMware
vSphere 4 white paper
.
2.1. FT Operations: Turning On and enabling
There are two types of FT operations that can be performed on a virtual machine: Turning FT on or off, and enabling or disabling FT.
The performance implications of these operations are as follows:
“Turn On FT”
prepares the virtual machine for FT.
• When FT is turned on, devices that are not supported with FT are prompted for removal, and the virtual machines memory
reservation is set to its memory size to prevent ballooning or swapping.
• Use of processor’s hardware MMU feature (AMD RVI/Intel EPT) results in non-determinism and therefore it is not supported
with FT. When FT is turned on for a virtual machine, hardware MMU feature is disabled for that virtual machine. However, virtual
machines that don’t have FT turned on can take advantage of hardware MMU on the same host.
• Turning on FT will not succeed if the virtual machine is powered on and is using Hardware MMU. In this case, the virtual machine
first needs to be either powered off, or migrated to a host that does not have hardware MMU. Similarly turning off FT on a
powered-on virtual machine would not make the virtual machine automatically use hardware MMU; the virtual machine would
need to be powered off and powered back on or migrated to a host that supports hardware MMU for the changes to take effect.
Please see
KB article 1008027
for more information on which guest OS and CPU combination requires power on/off operations
for changes to take effect.
“enable FT”
operation enables Fault Tolerance by live-migrating the virtual machine to another host to create a secondary
virtual machine.
• Since live-migration is a resource-intensive operation, limiting the frequency of enable/disable FT operations is recommended.
• The secondary virtual machine uses additional resources on your cluster. Therefore if the cluster has insufficient CPU or memory
resources, the secondary will not be created.
When “Turn on FT” operation succeeds for a virtual machine that is already powered on, it automatically creates a new secondary
virtual machine. So it has the same effect as “Enabling FT”.
2.2. resource Consumption
The additional resource requirements for running a virtual machine with Fault Tolerance enabled are as follows:
• CPU cycles and memory for running the secondary virtual machine
• CPU cycles for recording on the primary host and replaying on the secondary host
• CPU cycles for sending FT logging traffic from the primary host and receiving it on the secondary
• Network bandwidth for the FT logging traffic
Record and replay may consume different amounts of CPU depending on the event being recorded and replayed, and as a result,
slight differences in the CPU utilization of the primary and the secondary virtual machines is common and can be ignored.
2.3. Secondary Virtual Machine execution Speed
As explained in section 1.3, the hypervisor may slow down the primary virtual machine if the secondary is not keeping up pace with
the primary. Secondary virtual machine execution can be slower than the primary for a variety of reasons:
• The secondary host has a CPU with a significantly lower clock frequency
• Power management is enabled on the secondary host, causing the CPU frequency to be scaled down
• The secondary virtual machine is contending for CPU with other virtual machines
