Adaptive message passing polling for energy efficiency: Application to software‐distributed shared memory over heterogeneous computing resources

Abstract

Autonomous vehicles, smart manufacturing, heterogeneous systems, and new high-performance embedded computing (HPEC) applications can benefit from the reuse of code coming from the high-performance computing world. However, unlike for HPC, energy efficiency is critical in embedded systems, especially when running on battery power. Code base from HPC mostly relies on the message passing interface (MPI) message passing runtime to deal with distributed systems. MPI has been designed primarily for performance and not for energy efficiency. One drawback is the way messages are received, in an energy-consuming busy-wait fashion. In this work, we study a simple approach in which receiving processes are put to sleep instead of constantly polling. We implement this strategy at the user level to be transparent to the MPI runtime and the application. Experiments are conducted with OpenMPI, MPICH, and MPC, using a video processing application and a software-distributed shared memory system deployed over two heterogeneous platforms, including the Christmann RECS|Box Antares Microserver. Results show significant energy savings. In some particular cases involving process colocation, we also observe better performance using our strategy which can be explained by a better sharing of the computing resource.

Software-Distributed Shared Memory for Heterogeneous Machines: Design and Use Considerations

“…It has been introduced to limit the energy consumption that occurs when polling for new messages, as it is designed in most of the MPI runtimes. More information about the micro-sleeping mechanism can be found in this article [8]. From the time decomposition, Sleep and user code times can be interpreted as an efficient use of resources, while Sync MP and S-DSM code times can be considered as overhead.…”