The management system for the research cloud has beened “scaled out” from 1 node to 3 nodes.  This should improve performance and reliablity as well as allowing future updates to be applied without service interruption.