The configuration is as follows:

• VirtualService objects, which define weight distributions between

requests sent to different subsets of a service

• DestinationRule objects, which define the version of the underlying

Pods that a subset is defined by – in other words, to which the traffic

is routed