Open jcarreira opened 7 years ago
Right now the RDMA backend supports two methods for dealing with memory pinning:
RDMAMem
RDMAClient
Currently, mostly version 2 is being used which is very expensive.
A potentially more efficient alternative would be to
The externall allocator from boost doesn't seem very efficient when compared against libc's.
It might still be faster to pre-pin+BoostAllocator than to pin memory every time.
Right now the RDMA backend supports two methods for dealing with memory pinning:
RDMAMem
) previously pinned is passed to theRDMAClient
to avoid having to register memoryRDMAMem
pointer is passed, Cirrus registers the memory region where the data passed to Cirrus livesCurrently, mostly version 2 is being used which is very expensive.
A potentially more efficient alternative would be to