BIER Fast ReRoute

specifies "Bit Index Explicit Replication" (BIER). It provides optimal forwarding of multicast data packets through a "multicast/BIER domain". It does not require the use of a protocol for explicitly building multicast distribution trees, and it does not require intermediate nodes to maintain any per-flow state. proposes a tunnel-based fast re-route (FRR) method for protecting a node or link in the core of a BIER domain, which is called tunnel-based BIER-FRR. It tunnels BIER packets around the failure to BIER nodes downstream in multicast distribution trees. For a (next hop) node failure, it tunnels BIER packets to the next next hop nodes (NNHs). The BIFT in every BFR is enhanced to have two forwarding entries for every BFER. One is the primary forwarding entry with primary NH such as BFR neighbor and primary bit mask, and the other is the backup forwarding entry with backup NH such as NNH and backup bit mask. Using one BIFT in a BFR for both normal and backup forwarding will save memory. In normal operations, the primary forwarding entries are used to forward BIER packets. When a failure such as a node failure happens, the backup forwarding entry corresponding to the failure and the other primary forwarding entries are used to forward BIER packets. In the BIFT, the primary bit mask in every primary forwarding entry is computed before the failure. After the failure, the primary bit mask needs to be recomputed from the changed topology. Before the primary bit mask is recomputed and updated, some of BIER packets may be forwarded incorrectly. This document describes a mechanism for fast re-route (FRR) protection against the failure of a node or link in the core of a BIER domain, which resolves the above issue. It is based on LFA, which is called LFA-based BIER-FRR. On a BFR, there is a FRR BIFT for each of its neighbors, which has considered the neighbor failure. There is one forwarding entry for every BFER in any BIFT, including normal BIFT and FRR BIFT. This may use more memory. In normal operations, the normal BIFT is used to forward BIER packets. When a neighbor fails, the BFR as PLR uses the FRR BIFT for the neighbor to forward BIER packets. For a BIER packet to traverse the BFR and the failed neighbor, the BFR reroutes the packet around the failed neighbor using the FRR BIFT for the neighbor. For a BIER packet to traverse the BFR and any other neighbors, the BFR forwards the packet to its expected next hop neighbors using the forwarding entries with these BFR neighbors in the FRR BIFT.

Bit-Forwarding Router. Bit-Forwarding Ingress Router. Bit-Forwarding Egress Router. BFR Identifier. It is a number in the range [1,65535]. BFR Neighbor. Forwarding Bit Mask. An IP address (either IPv4 or IPv6) of a BFR. Bit Index Routing Table. It is a table that maps from the BFR-id (in a particular sub-domain) of a BFER to the BFR-prefix of that BFER, and to the BFR-NBR on the path to that BFER. Bit Index Forwarding Table. Fast Re-Route. Point of Local Repair. Loop-Free Alternate. Remote LFA. Remote LFA with Directed forwarding. Interior Gateway Protocol. Link State DataBase. Shortest Path First. Shortest Path Tree. The SPT rooted at node R using LSDB before X fails (i.e., old LSDB). The SPT rooted at node R using LSDB without X after X fails (i.e., new LSDB). The set of nodes that are reachable from R without going through X. In other words, it is the set of nodes that are not downstream of X in SPT-old(R). The set of nodes that are reachable from R or a neighbor of R, without going through X. The set of nodes that do not use X to reach destination D using the old LSDB. A member of both the P-Space P(R, X) (or the extended P-Space P'(R, X)) and the Q-Space (D, X).

A Bit-Forwarding Router (BFR) in a BIER sub-domain builds and maintains a "FRR Bit Index Routing Table" (FRR-BIRT) for each of its BFR Neighbors (BFR-NBRs) to provide BIER-FRR. The BFR builds each FRR-BIRT based on a BIRT defined in . A "FRR Bit Index Forwarding Table" (FRR-BIFT) is derived from a FRR-BIRT in the same way as a BIFT is derived from a BIRT, which is defined in . The forwarding procedure defined in is enhanced/updated for FRR-BIFTs. Once the BFR as a PLR detects the failure of its BFR-NBR X, it uses the FRR-BIFT for X to forward packets with BIER headers to get around failed X according to the updated/enhanced forwarding procedure.

This section briefs the BIRT, BIFT and forwarding procedure defined in . There is a "Bit Index Routing Table" (BIRT) for a BIER sub-domain on a BFR. The BIRT maps the BFR Identifier (BFR-id) (in the sub-domain) of a Bit-Forwarding Egress Router (BFER) to the BFR-prefix of that BFER, and to the BFR-NBR on the shortest path to that BFER. In other words, the BIRT has a route or say a next hop (i.e., BFR-NBR on the path) to every BFER. From the BIRT on the BFR, a "Bit Index Forwarding Table" (BIFT) is derived. In addition to having a route to a BFER in each row of the BIFT which is the same as the BIRT, it has a "Forwarding Bit Mask" (F-BM) in its each row. For the rows in the BIRT that have the same SI and the same BFR-NBR, the F-BM for each of these rows in the BIFT is the logical OR of the BitStrings of these rows. This BIFT is programmed into the data plane and used to forward a packet with a BIER header. The header contains SI, BitString, BitStringLength, and sub-domain. When a BFR receives a packet, for each BFER k (from the rightmost to the leftmost) represented in the SI and BitString of the packet, if BFER k is the BFR itself, the BFR copies the packet, sends the copy to the multicast flow overlay and clears bit k in the original packet; otherwise the BFR finds the row (i.e., forwarding entry) in the BIFT for the sub-domain using the SI and BitString as the key or say index, and then copies, updates and forwards the packet to the BFR-NBR (i.e., the next hop) indicated by the row (i.e., forwarding entry). After copying the packet and before forwarding it to the BFR-NBR, the packet's BitString is updated by ANDing it with the F-BM in the forwarding entry (i.e., PacketCopy->BitString &= F-BM). After forwarding the updated packet to a BFR-NBR and before forwarding the original packet to another BFR-NBR, the original packet's BitString is changed by ANDing it with the INVERSE of the F-BM (i.e., Packet->BitString &= ~F-BM).

Each BFR in a BIER sub-domain builds and maintains a number of "FRR Bit Index Routing Tables" (FRR-BIRTs). There is a FRR-BIRT for each BFR-NBR of the BFR. The BFR builds each FRR-BIRT based on its BIRT. It has the same format as the BIRT. The FRR-BIRT for BFR-NBR X of the BFR considers the failure of X and maps the BFR-id (in the sub-domain) of a BFER to the BFR-prefix of that BFER, and to BFR-NBR N on the path to that BFER. In other words, the FRR-BIRT has a route or say a next hop (i.e., BFR-NBR N on the path, where N is not X) to every BFER when BFR-NBR X fails. The BFR may build the FRR-BIRT for BFR-NBR X by copying its BIRT to the FRR-BIRT first, and then change the next hop with value BFR-NBR X in the FRR-BIRT to a backup next hop (BNH) to protect against the failure of X. In other wards, for the BFR-id of a BFER in the FRR-BIRT for BFR-NBR X, if the next hop BFR-NBR on the path to the BFER is X, it is changed to a BNH when there is a BNH on a backup path to the BFER without going through X and the link from the BFR to X. If there is not any BNH to a BFER to protect against the failure of X, the next hop BFR-NBR X to the BFER in the FRR-BIRT for BFR-NBR X is changed to NULL. For a multicast packet having the BFER as one of its destinations, if the next hop BFR-NBR to the BFER is NULL, the BFR does not send the packet to the next hop BFR-NBR NULL but drops it when X fails. Note: In another option, the next hop BFR-NBR X to the BFER in the FRR-BIRT for BFR-NBR X keeps unchanged when there is not any BNH to the BFER to protect against the failure of X. In this case, for a multicast packet having the BFER as one of its destinations, the BFR sends the packet to X when X fails. In one implementation, the BNH is the Loop-Free Node-Protecting Alternate defined in to protect against the failure of X and link from the BFR to X. In another implementation, the BNH is the virtual Loop-Free Alternate (LFA), i.e., PQ node, defined in . In a special case, a PQ node is a Loop-Free Node-Protecting Alternate defined in .

From each FRR-BIRT on the BFR, a "FRR Bit Index Forwarding Table" (FRR-BIFT) is derived. In addition to having a route to a BFER in each row of the FRR-BIFT which is the same as the FRR-BIRT, it has a "Forwarding Bit Mask" (F-BM) in its each row. For the rows in the FRR-BIRT that have the same SI and the same BFR-NBR, the F-BM for each of these rows in the FRR-BIFT is the logical OR of the BitStrings of these rows. This FRR-BIFT is programmed into the data plane and is not used to forward any packet in normal operations. It is activated to forward a packet with a BIER header once the BFR detects the failure of BFR-NBR. The header contains SI, BitString, BitStringLength, and sub-domain.

The forwarding procedure defined in is updated/enhanced for a FRR-BIFT to consider the case where the next hop BFR-NBR to a BFER is NULL. For a multicast packet with the BitString indicating a BFER as one of its destinations, the updated forwarding procedure checks whether the next hop BFR-NBR to the BFER in the FRR-BIFT is NULL. If it is NULL, the procedure will not send the packet to this next hop BFR-NBR NULL but drop the packet. The updated procedure is described in . It is used with a FRR-BIFT for BFR-NBR X on a BFR to forward multicast packets when X fails. It can also be used with a BIFT on the BFR to forward multicast packets in normal operations.

The FRR-BIFTs will be pre-computed and installed ready for activation when a failure is detected. Once the BFR detects the failure of its BFR-NBR X, it activates the FRR-BIFT for X to forward packets with BIER headers and de-activates its BIFT. After activation of the FRR-BIFT, it remains in effect until it is no longer required. In general, when the routing protocol has re-converged on the new topology taking into account the failure of X, the BIRT is re-computed using the updated LSDB and the BIFT is re-derived from the BIRT. Once the BIFT is installed ready for activation, it is activated to forward packets with BIER headers and the FRR-BIFT for X is de-activated. From the new topology, the BFR computes/re-computes the FRR-BIRT for each BFR-NBR Y of the BFR and the FRR-BIFT for Y is derived/re-derived from the FRR-BIRT for Y. The FRR-BIFT is installed/re-installed ready for activation when Y fails.

This section illustrates an example application of BIER FRR on a BFR in a BIER topology in .

An example BIER topology for a BIER sub-domain is shown in . It has 8 nodes/BFRs A, B, C, D, E, F, G and H. Each of the links connecting these nodes/BFRs has a cost. The link cost of 1 is default and is not indicated in the figure. The link cost of other value such as 2 is indicated in the figure.

Nodes/BFRs D, F, E, H and A are BFERs and have BFR-ids 1, 2, 3, 4, and 5 respectively. For simplicity, these BFR-ids are represented by (SI:BitString), where SI = 0 and BitString is of 5 bits. BFR-ids 1, 2, 3, 4, and 5 are represented by (0:00001), (0:00010), (0:00100), (0:01000) and (0:10000) respectively.

Every BFR in a BIER sub-domain/topology builds and maintains a Bit Index Routing Table (BIRT). For the BIER topology in , each of 8 nodes/BFRs A, B, C, D, E, F, G and H builds and maintains a BIRT using the LSDB for the topology. The BIRT built on BFR B (i.e. node B) is shown in .

The 1st row in the BIRT indicates that the next hop BFR-NBR on the shortest path to BFER D with BFR-id 1 is BFR C. The 2nd row in the BIRT indicates that the next hop BFR-NBR on the shortest path to BFER F with BFR-id 2 is BFR C. The 3rd row in the BIRT indicates that the next hop BFR-NBR on the shortest path to BFER E with BFR-id 3 is BFR E. The 4-th row in the BIRT indicates that the next hop BFR-NBR on the shortest path to BFER H with BFR-id 4 is BFR C. The 5-th row in the BIRT indicates that the next hop BFR-NBR on the shortest path to BFER A with BFR-id 5 is BFR A. From this BIRT on BFR B, a Bit Index Forwarding Table (BIFT) is derived. This BIFT is shown in . The 1st, 2nd and 4-th rows in the BIRT have the same SI = 0 and next hop BFR-NBR = C. The F-BM for each of these three rows in the BIFT is the logical OR of the BitStrings of these rows, which is 01011 (00001 OR 00010 OR 01000 = 01011). The F-BM for 3rd row in the BIFT is 00100. The F-BM for 5-th row in the BIFT is 10000.

Every BFR in a BIER sub-domain/topology builds and maintains a number of FRR Bit Index Routing Tables (FRR-BIRTs). For the BIER topology in , each of 8 nodes/BFRs A, B, C, D, E, F, G and H builds and maintains a number of FRR-BIRTs using the LSDB for the topology for its every BFR-NBR. For example, BFR B (i.e., node B) in the BIER topology builds and maintains four FRR-BIRTs for its four BFR-NBRs (BFR C, BFR E, BFR A and BFR G) respectively. The FRR-BIRT for BFR C built by BFR B is shown in .

The 1st row in the FRR-BIRT indicates that the next hop BFR-NBR on the path to BFER D with BFR-id 1 is BFR G. G is the Loop-Free Node-Protecting Alternate defined in to protect against the failure of C and link from B to C. The 2nd row in the FRR-BIRT indicates that the next hop BFR-NBR on the path to BFER F with BFR-id 2 is BFR E. E is the Loop-Free Node-Protecting Alternate defined in to protect against the failure of C and link from B to C. The 3rd row in the FRR-BIRT indicates that the next hop BFR-NBR on the path to BFER E with BFR-id 3 is BFR E. The 4-th row in the FRR-BIRT indicates that the next hop BFR-NBR on the path to BFER H with BFR-id 4 is BFR G. G is the Loop-Free Node-Protecting Alternate defined in to protect against the failure of C and link from B to C. The 5-th row in the FRR-BIRT indicates that the next hop BFR-NBR on the path to BFER A with BFR-id 5 is BFR A. From this FRR-BIRT for BFR C on BFR B, a FRR Bit Index Forwarding Table (FRR-BIFT) is derived. This FRR-BIFT for BFR C is shown in . The 1st and 4-th rows in the FRR-BIRT have the same SI = 0 and next hop BFR-NBR = G. The F-BM for each of these two rows in the FRR-BIFT is the logical OR of the BitStrings of these rows, which is 01001 (00001 OR 01000 = 01001).

The 2nd and 3rd rows in the FRR-BIRT have the same SI = 0 and next hop BFR-NBR = E. The F-BM for each of these two rows in the FRR-BIFT is the logical OR of the BitStrings of these rows, which is 00110 (00010 OR 00100 = 00110). The F-BM for 5-th row in the FRR-BIFT is 10000. The number of entries in a FRR BIFT is the number of BFERs. Each FRR BIFT on a BFR can be compressed through combining all the entries with the same BFR-BNR and F-BM into one entry. The number of entries in a compressed FRR BIFT is the number of neighbors of the BFR minus one. For example, the compressed FRR-BIFT for BFR C on BFR B is shown in . The number of entries in it is three, which equals the number (four) of neighbors of BFR B minus one.

For a BIER packet with a BFR-ID as a destination, the entry containing the BFR-ID is used to forward the packet.

Suppose that there is a multicast traffic from BFR A as ingress/BFIR to egresses/BFERs D, F, E and H. For every packet of the traffic, after receiving it, BFR A adds a BIER header into the packet and sends the packet with the BIER header to BFR B. The BIER header contains (SI:BitString) = (0:01111) for egresses/BFERs D, F, E and H. In normal operations, after receiving the packet from BFR A, BFR B copies, updates and sends the packet to BFR C and BFR E using the BIFT on BFR B according to the forwarding procedure defined in . Once BFR B detects the failure of its BFR-NBR X, after receiving the packet from BFR A, BFR B copies, updates and sends the packet using the FRR-BIFT for X on BFR B to avoid X and link from B to X according to the forwarding procedure defined in . For example, once BFR B detects the failure of its BFR-NBR C, after receiving the packet from BFR A, BFR B copies, updates and sends the packet to BFR G and BFR E using the FRR-BIFT for BFR C on BFR B to avoid C and link from B to C. The packet received by BFR B from BFR A contains (SI:BitString) = (0:01111). The rightmost one bit in BitString is bit 1. For BFER 1 (0:00001) (i.e., BFR D as BFER), BFR B gets the 1st row (i.e., forwarding entry) in the FRR-BIFT for BFR C. The next hop BFR-NBR is G in the row. BFR B copies, updates and forwards the packet to G. The packet sent to G contains the updated BitString = 01001, which is 01111 & F-BM in the row = 01111 & 01001. After sending the packet to G, BFR B updates the original packet by ANDing its BitString with the INVERSE of the F-BM in the row. The updated BitString = 00110, which is 01111 & ~F-BM in the row = 01111 & 00110. For the packet containing BitString = 00110, the rightmost one bit in BitString is bit 2. For BFER 2 (0:00010) (i.e., BFR F as BFER), BFR B gets the 2nd row (i.e., forwarding entry) in the FRR-BIFT for BFR C. The next hop BFR-NBR is E in the row. BFR B copies, updates and forwards the packet to E. The packet sent to E contains the updated BitString = 00110, which is 00110 & F-BM in the 2nd row = 00110 & 00110. After sending the packet to E, BFR B updates the original packet by ANDing its BitString with the INVERSE of the F-BM in the 2nd row. The updated BitString = 00000, which is 00110 & ~F-BM in the row = 00110 & 11001. The updated packet has BitString without any one bit. BFR B finishes forwarding the packet from A to D, F, E and H.

TBD.

No requirements for IANA.

The authors would like to Jeffrey Zhang, Daniel Merling and Geng Xuesong for their comments to this work.