Hi. Do you see performance changes between file reads/writes? (can you check the performance of your code without doing the memory registration? to check if the algorithm that you are using fro sending data hurts the performance)
thanks Dotan yangdong wrote: > Hello! I implemented a message passing subsystem -- AMP for our cluster > file system--DCFS3, > I want AMP to run over Infiniband, I designed it in kernel(verbs). > when client writing file, my protocol is implemented by RDMA WRITE, and > reading file by RDMA > READ, but perf of AMP is less than my expectation. Single thread (1t) > read is 215MB/s, 4t is 340MB/s. > And I see that the read file perf of NFS/RDMA is close to 600MB/s (1t). > I made use of FMR and All physical reg ..and scatter/gather, but why the > perf of AMP is lower? > > > > _______________________________________________ general mailing list [email protected] http://lists.openfabrics.org/cgi-bin/mailman/listinfo/general To unsubscribe, please visit http://openib.org/mailman/listinfo/openib-general
