Abstract
Routing table lookup, or its memory latency, is the most serious bottleneck of high performance routing. Flow-wise or group-wise routing table entries for QoS guaranteed or multicast communications need a large routing table, which further increase the latency. Recent routers have multiple input interfaces and output interfaces connected by a high bandwidth cross connect. Such architecture allows parallel routing table look up at each interface but the number of interfaces limits the parallelism. By having two cross connects, one between input interfaces and routing engines, the other between the routing engines and output interfaces, it is possible to make the routing table look up unlimitedly parallel. The remaining problem is how to decide the proper routing engine at the input interface. Hash parallel routing is a technique to use the hashed value of the destination addresses of packets to decide the routing engine. Label parallel routing is a technique to decide the routing engine at the time of signaling and ask the router at the previous hop to put the decision result in the link label.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.