[openib-general] RMPP implementation
mshefty at ichips.intel.com
Wed Nov 17 09:07:48 PST 2004
Fab Tillier wrote:
>>1. Issue one send at a time. Additional sends are not transfered until
>> the first send completes.
> Isn't #1 the simplest to implement? Turnaround on the send queue should be
> pretty quick, so send performance should be fine. I say do whatever is
> simplest, and then optimize from there, and to me that means #1 at the
> moment. What are the reasons to *not* do #1?
It's simpler to implement, and would definitely be the easiest to do on
redirected QPs. The only disadvantage is that it lowers the throughput
between two clients. Also, this is a relatively small decrease in
complexity with respect to the rest of RMPP.
A couple of other areas that will need to be addressed include: RMPP
timeouts, receive window sizes, and user-mode support.
More information about the general