FPGA BASED SOFTWARE TESTING PRIORITIZATION USING RnK-MEANS CLUSTERING

Testing the software is to validate its correctness when it is deployed in its actual environment. Various test cases should be implemented and tested to validate the software. When more than one test case is involved, the order of testing needs to be prioritized to optimize the testing process. This paper proposed a prioritization method with repeated n times K means (RnK-means) clustering. Priority for the test cases is assigned based on the cluster mean values by executing RnK-means for each factor of test cases. Existing techniques are calculating merely the average of factor weights for each test case for deciding priority. The proposed method involves K-means computations and it is accelerated by FPGA for deciding priority. The observed results proved 20 percent better performance with RnK-means clustering than the existing weighted average method.