2 simple parallel distribution question

Consider a completely connected network of p nodes. For the four communication operations in Problem 4.25 derive an expression for the parallel run time of the hypercube algorithms on the completely connected network. Comment on whether the added connectivity of the network yields improved performance for these operations.

Consider Dijkstra’s single-source shortest-path algorithm described in Section 6.6.9. Describe why a column-wise distribution is preferable to a row-wise distribution of the weighted adjacency matrix.