<?xml version="1.0" encoding="UTF-8"?>
<reference anchor="I-D.xu-idr-fare-in-mpson" target="https://datatracker.ietf.org/doc/html/draft-xu-idr-fare-in-mpson-01">
   <front>
      <title>Fully Adaptive Routing Ethernet in Multi-Plane Scale-Out Networks</title>
      <author initials="X." surname="Xu" fullname="Xiaohu Xu">
         <organization>China Mobile</organization>
      </author>
      <author initials="K." surname="Patel" fullname="Keyur Patel">
         <organization>Arrcus</organization>
      </author>
      <author initials="H." surname="Wang" fullname="Haibo Wang">
         <organization>Huawei Technologies</organization>
      </author>
      <author initials="W." surname="Sun" fullname="Wei Sun">
         <organization>Yunsilicon Technology</organization>
      </author>
      <author initials="G." surname="Ma" fullname="Guoqiang Ma">
         <organization>NebulaMatrix</organization>
      </author>
      <date month="August" day="16" year="2026" />
      <abstract>
	 <t>   FARE-BGP enables weighted ECMP load balancing using a path-bandwidth
   extended community.  FARE-in-SUN extends this mechanism from switches
   to GPUs for scale-up networks, which are typically multi-plane.
   Large AI training clusters are increasingly adopting multi-plane
   scale-out network topologies.  This document further extends FARE-BGP
   from switches to RoCE NICs (RNICs) for such multi-plane scale-out
   networks.  The document also presents two techniques to address route
   scalability concerns caused by the injection of numerous host routes.


	 </t>
      </abstract>
   </front>
   <seriesInfo name="Internet-Draft" value="draft-xu-idr-fare-in-mpson-01" />
   
</reference>
