Browse Prior Art Database

BIT INDEX EXPLICIT REPLICATION (BIER) FORWARDING WITH PER-ENTROPY BIT-STRINGS

IP.com Disclosure Number: IPCOM000244899D
Publication Date: 2016-Jan-27
Document File: 4 page(s) / 21K

Publishing Venue

The IP.com Prior Art Database

Related People

Neale Ranns: AUTHOR [+7]

Abstract

Per-entropy value forwarding masks per- Bit Index Explicit Replication Forwarding Router Neighbor are used to eliminate the wasted processing time caused by dropping replicated packets late in the forwarding cycle, thus increasing line-card overall throughput.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 53% of the total text.

Page 01 of 4

       BIT INDEX EXPLICIT REPLICATION
(BIER) FORWARDING WITH PER-ENTROPY BIT-STRINGS

AUTHORS:

  Neale Ranns Ijsbrand Wijnands Greg Shepherd

   Samuel Liu Gaofeng Tao Alan Xiao-Rong Wang

Lianxiang Wang

CISCO SYSTEMS, INC.

ABSTRACT

    Per-entropy value forwarding masks per- Bit Index Explicit Replication Forwarding Router Neighbor are used to eliminate the wasted processing time caused by dropping replicated packets late in the forwarding cycle, thus increasing line-card overall throughput.

DETAILED DESCRIPTION

    "Bit Index Explicit Replication" (BIER) is an architecture that performs forwarding of multicast data packets through a "multicast domain" without requiring the use of a protocol for explicitly building multicast distribution trees, and without requiring intermediate nodes to maintain any per-flow state.

    A multicast data packet enters a BIER domain at a "Bit-Forwarding Ingress Router" (BFIR), and leaves the BIER domain at one or more "Bit-Forwarding Egress Routers" (BFERs). A BFR that receives a multicast data packet from another BFR in the same BIER domain, and forwards the packet to another BFR in the same BIER domain, will be known as a "transit BFR" for that packet. A single BFR may be a BFIR for some multicast traffic while also being a BFER for some multicast traffic and a transit BFR for some multicast traffic. In fact, a BFR may be the BFIR for a given packet and may also

Copyright 2016 Cisco Systems, Inc.

1


Page 02 of 4

be (one of) the BFER(s), for that packet; it may also forward that packet to one or more additional BFRs.

    When forwarding BIER packets, there are two reasons why a replicated packet copy would be dropped on the egress card for a given flow:

1. Early--when the logical AND result of the packet's bit-string with the replicated to neighbor's forwarding mask yields zero.

2. Late--when after the above yields non-zero, and forwarding continues by traversing the unicast forwarding chain for the neighbor, and where making a load-balancing (LB) decision that results in an adjacency/output port that is not local to the line-card (LC)/ASIC.

    A late drop thus consumes more resources than the early drop. Avoid the late drop is an important problem to solve.

    The load balancing decision on the egress card is between parallel links to the router's directly connected BFR-Neighbor (BFR-NBR). This is a route the router learned via IGP. It is therefore non-recursive, meaning in terms of load balancing decisions, there will be only one.

    A router has a limit on the number of choices it can have in a load balancing decision. The maximum for XR platforms is 64. Hash functions are ultimately modulo arithmetic: hash = entropy % num_choices. The above const...