[libfabric-users] Connection-based vs Connection-less

James Board jpboard2 at yahoo.com
Thu Sep 12 06:06:56 PDT 2019

I have somesimple  newbie questions about libfabrics.  In the context of traditional networking, I think I understand connection-based vs connection-less communication fairly well.  Connection-based implies that the route between A and B is configured before any data communication occurs, and that route stays persists through multiple messages.  Connection-less implies that each message finds it's own route through the network at the time the message is sent, and that route may change from message to message.

My first question is whether the above description is true for libfabrics connections as well?  Are there other key distinctions between connection-based vs connection-less communication?

My second question is about the address vector (AV).  In libfabrics, the AV is required for connection-less, but is not used at all for connection-based.  Is that correct?  What exactly does the address vector do?  What is it's function?

Finally, in the fabtests examples, I looked at rdm_rma_simple.c (along with shared.c) and pruned it to get rid of the unnecessary code.  I still can't figure out if rdm_rma_simple.c is connection-less or connection-based.  I think it's connection-less, but there is a lot of code to setup the connection beforehand.  Is it indeed connection-less?

Assuming rdm_rma_simple.c is connection-less, which of the fabtests would be a simple connection-based example?  I want to start out with the simplest fabtests code that is connection-based.

Thanks in advance
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openfabrics.org/pipermail/libfabric-users/attachments/20190912/6620288e/attachment.html>

More information about the Libfabric-users mailing list