Network Performance Aware MPI Collective Communication Operations in the Cloud
In this paper, the authors examine the performance of collective communication operations in Message Passing Interfaces (MPI) in the cloud computing environment. The awareness of network topology has been a key factor in performance optimizations for existing MPI implementations. However, virtualization in the cloud environment not only hides the network topology information from the users, but also causes traffic interference and dynamics to network performance. Existing topology-aware optimizations are no longer feasible in the cloud environment. Therefore, they develop novel network performance aware algorithms for a series of collective communication operations including broadcast, reduce, gather and scatter. They further implement two common applications, N-body and Conjugate Gradient (CG).