SMB Direct Implementation via Linux Device Driver for Samba Design Doc, final?
Or Gerlitz
ogerlitz at mellanox.com
Mon Aug 26 05:09:20 MDT 2013
On 26/08/2013 10:42, Stefan (metze) Metzmacher wrote:
>> >Another approach would be to have a single process responsible for all RDMA
>> >handling and have the smbd's communicate with that process about new incoming
>> >connections and reads and writes. While this would work, and could eliminate
>> >multiple copies of the data with shared memory, it would involve a context
>> >switch for most if not all RDMA transfers.
> Only reason for doing this would be research how the protocol works...
>
Yes, this is possible too. AFAIK, there another cluster file system rdma
implementation which uses
this practice, of daemon process which handle RDMA and uses
shared-memory with other processes.
In that respect, I am not sure to follow your comment on "could
eliminate multiple copies of the data with shared
memory" as I understood the design document you have sent, the intention
is to allocate memory in the kernel
rdma char-device which will be
1) used for send/recv/rdma
2) mmap-ed to user space
So where are the copies in this design?
As for "involve a contextswitch for most if not all RDMA transfers",
indeed, in this architecture you
will have contextswitch between the rdma daemon to the worker processes
vs. user/kernel mode switch
when you have each worker opening the rdma char-device.
Or.
More information about the samba-technical
mailing list