On the Resiliency of Static Forwarding Tables

Abstract

Fast Reroute (FRR) and other forms of immediate failover have long been used to recover from certain classes of failures without invoking the network control plane. While the set of such techniques is growing, the level of resiliency to failures that this approach can provide is not adequately understood. In this paper, we embarked upon a systematic algorithmic study of the resiliency of forwarding tables in a variety of models (i.e., deterministic/probabilistic routing, with packet-header-rewriting, with packet-duplication). Our results show that the resiliency of a routing scheme depends on the "connectivity" k of a network, i.e., the minimum number of link deletions that partition a network. We complement our theoretical result with extensive simulations. We show that resiliency to 4 simultaneous link failures, with limited path stretch, can be achieved without any packet modification/duplication or randomization. Furthermore, our routing schemes provide resiliency against k−1 failures, with limited path stretch, by storing log(k) bits in the packet header, with limited packet duplication, or with randomized forwarding technique.

Distributed Consistent Network Updates in SDNs: Local Verification for Global Guarantees

“…For destination-based routing, observe that we do not need to forward to a node with a depth exactly one smaller, but any smaller depth (or higher version) would suffice. In this context, fault-tolerance could benefit benefit from link-disjoint forwarding trees [15], which can be computed efficiently [16], along with appropriate optimization for route lengths [17,18].…”

CASA: Congestion and Stretch Aware Static Fast Rerouting

“…For routing decisions we assume that forwarding rules can match packet header fields as well as the in-port (the port from which a packet arrives at v, already used in [15] for fast reroute), and depending on this match, define the outgoing port to which a packet is forwarded at v. In other words, the focus of this paper is on oblivious (i.e., static) routing algorithms which do not rely on any dynamic state at nodes (e.g., counters) or in packets: we do not allow packet tagging. While tagging can improve the robustness of routing [16], [17], it is often undesirable in practice to change header fields.…”

“…In particular, thanks to the possibility to match the in-port and the source of a packet, a node can be traversed multiple times during failover, without ending up in an infinite loop. That is, failover routes may not be simple paths but form walks, e.g., consider a network with a dead-end, forcing the packets to return along the same link [16].…”

“…For comparison, we also include the circular arborescence failover scheme from Chiesa et al [32] (also in [16]). In particular, this allows us to judge the practical impact of CASA's approach of combining BIBDs with circular arborescence routing.…”

Local Fast Rerouting with Low Congestion: A Randomized Approach

“…In [32], [33], Chiesa et al conjecture [32], [33] that is at least always possible to deterministically achieve what they call ideal resilience: to preserve connectivity in a k-(edge-)connected network if there are at most k − 1 link failures. Today, it is still unknown whether this conjecture holds.…”