MPEC (Matrix Multiplication Performance Estimator on Cloud)

The proposed algorithm predicts the latency incurred when executing distributed matrix multiplication tasks of various input sizes and shapes with diverse instance types and a different number of worker nodes on cloud computing environments. The NMF algorithm is widely used in recommendation systems. It factorizes an input sparse matrix A into two dense matrices W and H. The following table shows the latency of matrix multiplication when updating the factorized matrix W. Next tables represent three matrix multiplication scenarios in above mentioned it, respectively. In the table, the row shows the ec2 instance types and the columns is meaning of the number of Spark worker nodes.