Decoupling memory pinning from the application with overlapped on-demand pinning and MMU notifiers

Goglin, Brice

doi:10.1109/ipdps.2009.5160888

Cited by 1 publication

(1 citation statement)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Some work has also looked at the cost of memory registration in RDMA-enabled networks, especially its high costs for small buffers [11,23]. In a recent work presented in [24], researchers have proposed a pinning model in Open-MX based on the decoupling of memory pinning from the application, as a step toward a reliable pinning cache in the kernel. Their proposal enables full overlap of memory pinning with communication.…”

Section: Related Workmentioning

confidence: 99%

Exploiting application buffer reuse to improve MPI small message transfer protocols over RDMA-enabled networks

Rashti

Afsahi

2011

Cluster Comput

View full text Add to dashboard Cite

To avoid the memory registration cost for small messages in MPI implementations over RDMA-enabled networks, message transfer protocols involve a copy to intermediate buffers at both sender and receiver. In this paper, we propose to eliminate the send-side copy when an application buffer is reused frequently. We show that it is more efficient to register the application buffer and use it for data transfer. The idea is examined for small message transfer protocols in MVAPICH2, including RDMA Write and Send/Receive based communications, one-sided communications and collectives. The proposed protocol adaptively falls back to the current protocol when the application does not frequently use its buffers. The performance results over InfiniBand indicate up to 14% improvement for single message latency, close to 20% improvement for one-sided operations and up to 25% improvement for collectives. In addition, the communication time in MPI applications with high buffer reuse is improved using this technique.

show abstract

Section: Related Workmentioning

confidence: 99%