I would like to know if having set the configuration of one USRPN210 per Ethernet interface is better (in terms of minimizing the latency towards the Host) than the configuration of multiple USRPs connected to a host via a unique Eth interface (eg. 192.168.10.x )
Which config. have better performance ?
Please, for the config of one radio / Eth interface:
Is setting more than one Eth. interface means doing: ip addr add dev .... SubNet1(eg. 192.168.10.127/24)
and ip addr add dev .... SubNet2 (192.168.11.127/24)
and do I need to consequently change the IP @ of the second USRP N210 that will be communicating with the host via SubNet2 ?
On 05/02/2013 09:23 PM, NaceurElOuni wrote:
> and do I need to consequently change the IP @ of the second USRP N210 that
> will be communicating with the host via SubNet2 ?
> Best regards.
The IP protocol suite requires that every device have their own
address. Whether that device is a computer, or a toaster, or USRP, or a
or an iPhone, or your cyberdog, or whatever.
So, yes, every USRP will need its own address. Whether that address is
on a unique subnet (one ethernet card per USRP), or whether it's
shared, on the same subnet. That is a requirement of the way IP works.
Which means that you'll need to set unique addresses on each of your USRPs.
I encourage you to familiarize yourself with the entire doc tree