Check out the new USENIX Web site. next up previous
Next: Scenario 1 - Intranet Up: AVES Previous: Data Path Operations


Control Path Operations

The AVES control path mechanisms are used to configure the data path. That is, when a DNS query for a reusable-IP responder is processed by an AVES-aware DNS server, a waypoint is selected to virtualize the reusable-IP responder and the appropriate translation table entry is installed at the selected waypoint so it can correctly relay packets.

It is important to recall that the bindings between waypoint IP addresses and reusable-IP responders must be initiator-specific. This allows a waypoint IP address to simultaneously virtualize many reusable-IP responders, and as a result high connectivity can be achieved with a small number of IP addresses assigned to waypoints. Unfortunately, creating initiator-specific bindings is not easy because the IP address of the initiator is typically not available in a DNS query received by an AVES-aware DNS server. This is because, in practice, virtually all end host systems implement recursive DNS query [15]. That is, an end host sends a recursive DNS query to its local DNS server, and this local DNS server generates additional iterative queries on behalf of the end host, and eventually returns the answer to the end host. Thus, an AVES-aware DNS server generally only interacts with the local DNS server of the initiator, the IP address of the initiator is obscured. In the following, we describe two deployment scenarios and the appropriate techniques in each case to create initiator-specific bindings.



Subsections
next up previous
Next: Scenario 1 - Intranet Up: AVES Previous: Data Path Operations