Network surveillance (NS),which is basically a (partially or full )decentralized mode of detecting faulty and repaired status of distributed computing (DC)components,is a major part of RT fault-tolerant DC.One NS technique which is broadly applicable and amenable to rigorous quantitative anal ses of fault coverage is the supervisor-based network surveillance (SNS)scheme for use in point-to- point network based systems.Recently we have been studying an approach to implementing the SNS scheme within a middleware model,named the TMO (time-triggered message-triggered object)support middleware (TMOSM)which is layered above a COTS OS such as Windows 2000 and functions as a reliable RT execution engine for RT DC applications.The DC applications supported by the TMOSM are those designed by use of the TMO programming scheme.The implementation structure as well as the observed performance of a prototype implementation are discussed.