[nemo-dev] 答复: The connectivity expression in NEMO

Jizhigang jizhigang at huawei.com
Thu Aug 13 06:05:12 UTC 2015


I also support option 1. From the perspective of the user, option 1 is more simple, direct and easier to understand.
And I have a question. Based on option 2, A can communicate with C, but how to specify the bandwidth between them?

Regards,
Zhigang

发件人: nemo-dev-bounces at lists.opendaylight.org [mailto:nemo-dev-bounces at lists.opendaylight.org] 代表 Zhoutianran
发送时间: 2015年8月13日 11:51
收件人: nemo-dev at lists.opendaylight.org
主题: [nemo-dev] The connectivity expression in NEMO


Hi Team,



As I reviewed our model, I am thinking about this scenario.

I have the following user level topology, nodes are connected by connections:

[cid:image001.png at 01D0D5CF.E6E4B840]

There are two possible ways to interpret this as we have the same discussion on the IETF bar bof meeting.

1.  One node can only communicate with nodes with explicit connection. E.g, A can communicate with B and D, but not C.

2.  One node can communicate with all nodes with any connectivity. I.e, A can communicate with C since there is connectivity via B or D.



Personally, I support option 1. I think the “connection” should explicitly express the user’s intent for the connectivity. We cannot assume the intermediate nodes has the routing/forwarding capability.



However, option 2 will make the intent expression complex, and even worse have the user to know many details.

For example, with option 2, if the user only want A communicate with B and D. Then he has to know that C also has connectivity with A(this is hard). And he has to say “A cannot communicate with C”.





Best Regards,

Tianran
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.opendaylight.org/pipermail/nemo-dev/attachments/20150813/36224eaf/attachment-0001.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: image001.png
Type: image/png
Size: 10004 bytes
Desc: image001.png
URL: <http://lists.opendaylight.org/pipermail/nemo-dev/attachments/20150813/36224eaf/attachment-0001.png>


More information about the nemo-dev mailing list