Missing value imputation is crucial for the microarray data analysis since the missing values would degrade the performance of the downstream analysis, e.g. differentially expressed genes identification, gene clustering or classification. Although many missing value imputation algorithms has been proposed, convenient software tools are still lacking. The existing tools are not easy to use and cannot tell users how to choose the optimal imputation algorithm for their dataset. In this paper, we present an easy-to-use web server named IMDE (Impute Missing Data Easily). IMDE has two unique features. First, it provides much more missing value imputation algorithms than any existing tool. Second, it can suggest the optimal imputation algorithm for users' dataset after doing the performance evaluation.We used four different datasets to show that different optimal algorithms may be chosen for different datasets and for different selection schemes. We expect that IMDE will be a very useful server for solving the missing value problem in the microarray data. IMDE is freely available at http://cosbi.ee.ncku.edu.tw/IMDE/.