> > The MTU reported by "ifconfig ib0" is the MTU used by the Linux > TCP/IP network stack. The MTU reported by ibv_devinfo is the > MTU that the hardware is capable of sending. This is limited > to 4K by the Infiniband specification. The reason the network > stack can have a higher MTU is that ib_ipoib is using the RC > QP protocol to send IP messages larger than the hardware MTU. > If you use "datagram" mode for ib_ipoib, you will see that > the network stack MTU is limited to the hardware MTU - 4.
Thank you Ralph!!! As far as common applications taking advantage of ib_ipoib Does it help using RC QP with a higher MTU than the hardware MTU? Does an Application, which uses Sockets API, by default make use of ib_ipoib, if it is enabled ? Is there any essential difference between IPoIB and ib_ipoib, or is it just a matter of usage ? Thank you, Amit > > > > Also Is there a document where I can read in detail about IPoIB and > applications that benefit from them. > > In general I understand that Socket based applications can make use > of IPoIB for a better bandwidth, thought NOT for a better transport > latency. > > In short I am trying to understand the difference and advantage, for > an Application using "Ethernet NIC" vs "InfiniBand HCA(IPoIB enabled)", > apart from knowing that there is no advantage in terms of transport > latency. > > _______________________________________________ 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
