In this paper, point and interval estimation of stress-strength reliability based on upper \emph{record ranked set sampling} (\emph{RRSS}) from one-parameter exponential distribution are considered. Maximum likelihood estimator (\emph{MLE}) as well as the uniformly minimum variance unbiased estimator (\emph{UMVUE}) of stress-strength parameter are derived and their performance are studied. Also, some confidence intervals for stress-strength parameter based on upper \emph{RRSS} are constructed and then compared on the basis of a simulation study. Finally, a data set has been analyzed for illustrative purposes.