Feature selection is crucial in reducing the large dimensionality of datasets like those found in microarray data. Microarray consists of thousands of genes, but not all are relevant for cancer prediction. In this scenario, the objective is to efficiently minimize the number of genes selected while maintaining the prediction performance. Selecting just one feature selection algorithm is difficult because they each have different computational characteristics. Therefore, it’s crucial to find an effective way to combine them to achieve the best results. This study presents a feature rank aggregation approach on the distributed framework using Borda count to aggregate the ranks. The data is divided across the feature space, with aggregation conducted in parallel across all partitions. Classification models are used to evaluate the efficiency of the proposed technique on the top-selected features from the final rank list. This study utilized two publicly available microarray datasets, demonstrating the effectiveness of our proposed parallel aggregator over traditional sequential methods. Additionally, our approach showed improved performance across all statistical measures.