DISTRIBUTED COMPUTING ARCHITECTURE WITH DYNAMICALLY RECONFIGURABLE HYPERVISOR NODES

摘要

A method of dynamically reconfiguring a distributed computing architecture having a plurality of processing nodes, where each processing node hosts a respective plurality of virtual machines, includes detecting a fault condition on a first processing node, assessing the criticality of a software function performed by each of the respective virtual machines on the first processing node, and reassigning at least one of the plurality of virtual machines on the first processing node to a second processing node if the at least one virtual machine is deemed critical.