Hello Linda, The whole section of Cloud Service Virtualization requirement is out of scope of NVo3. Many content which you discussed is belong to this section, so you say 'whole' is not appropriate. I think, except 3.6.2, 3.6.4, 3.6.19, other are nvo3 relevant part. So, these sections(3.6.2,3.6.4,3.6.19) would be moved to another draft. Thank you for your comments. Section 3.1 (The Evolution Problems of the Logical Network Topology in VMMI environments):
I don’t think the following statement is valid: “Since a large number of VMs and their applications are running in the same Layer-2 domain, it (VM migration) may be very stressful from bandwidth utilization viewpont of the data center switching network.” Applications communicate with each other via IP addresses. Yes, some instances are placed in the same subnet. VM migration is not the main reason for bandwidth utilization. There are many reasons for (poor) bandwidth utilization. “In order to improve the bandwidth utilization, it is required to upgrade the load balancing capability of the network which has numerous ECMP between different points.” This “load balance” has nothing to do with VM mobility. About these comments, you could read reply from ytsun. And I have read your draft(draft-rekhter-nvo3-vm-mobility-issues-03).I quite agree with you about the content of section 3.4 (Optimal IP Routing): + Optimal routing of a VM's outbound traffic. This means that as a given VM moves from one server to another, the VM's default gateway should be in a close topological proximity to the ToR that connects the server presently hosting that VM. Obviously, you think that there is close relation between 'the Optimal routing of a VM's outbound traffic' and network topology, and this with our draft in Section 3.1 (The Evolution Problems of The Logical Network Topology in VMMI environments) want to express the same point of view, and yhe target is consistent. So, from your draft we can know, your draft of view is agree with our draft's point of view. Don't you then changed your mind, want to overthrow viewpoint of yourself? In fact, we are all from different perspective, discussing the same problem. Similarly, for data center fat tree network in multipath forward, TRILL can make good use of network bandwidth. But in fact, for TRILL, fat tree also is not the best topology, so in the control plane (NVO3 / TRILL) still exist promotion space. Richard --------------- 横看成岭侧成峰， 远近高低各不同。 不识庐山真面目， 只缘身在此山中。 ---------------
_______________________________________________ nvo3 mailing list [email protected] https://www.ietf.org/mailman/listinfo/nvo3
