One of the ambitions when designing the Stream Control Transmission Protocol was to offer a robust transfer of traffic between hosts. For this reason SCTP was designed to support multihoming, which presumes the possibility to set up several paths between the same hosts in the same session. If the primary path between a source machine and a destination machine breaks down, the traffic may still be sent to the destination, by utilizing one of the alternate paths. The failover that occurs when changing path is to be transparent to the application.

This paper describes the results from experiments concerning SCTP failover performance, which means the time between occurrence of a break on the primary path until the traffic is run smoothly on the alternate path. The experiments are performed mainly to verify the Linux Kernel implementation of SCTP (LK-SCTP) and is run on the Emulab platform. The results will serve as a basis for further experiments.

The experiments are performed in a network without concurrent traffic and in conclusion the results from the experiments correspond well to the values found in other studies and they are close to the theoretical best values. As expected the parameter Path.Max.Retrans has a great impact on the failover time. One observation is that the failover time and the max transfer time for a message are dependent upon the status in the network when the break of the primary path occurs.

The stream control transmission protocol (SCTP) is an important component in the ongoing evolution towards IP in the fixed and mobile telephone networks. It is the transport protocol being used in the ongoing deployment of IETFpsilas signaling transport (SIGTRAN) architecture for tunneling of traditional telephony signaling traffic over IP. Further SCTP represents an alternative for future SIP signaling traffic. Key to the success of SCTP is its ability to recover from network failures, in particular failed network paths. SCTP includes multihoming and a failover mechanism which should swiftly shift from a failed or unavailable network path to a backup path. However, several studies have shown that SCTPpsilas failover performance is dependent on factors both related to protocol parameters and network conditions. This paper complements these studies by providing a comprehensive evaluation of the impact of SACK delay under various traffic distributions. The results show a clear relation between the traffic distribution and the impact of the SACK delay on SCTP failover performance. Severe negative effects are observed for low intensity traffic composed of individual signaling messages. On the other hand, our results show limited impact of SACK delay for high intensity and bursty traffic. Furthermore, the results show a limited increase in network traffic by reducing the SACK delay at low traffic intensities and even less impact on network traffic at high traffic intensities. Based on these results we recommend a decrease of the SCTP SACK timer to a small value in signaling scenarios

The Stream Control Transmission Protocol (SCTP) has not only been selected as the signaling transport protocol of choice in IETF SIGTRAN, the architecture that bridges circuit-switched and IP-based mobile core networks, but also plays a pivotal role in SAE/LTE, the next-generation UMTS/HSPA networks. To meet the redundancy requirements of telecom signaling traffic, SCTP includes a failover mechanism that enables rerouting of traffic from an unreachable network path to a backup path. However, the recommendations provided by IETF on how to configure the SCTP failover mechanism to meet telecom signaling requirements are kept quite general and leave much of the tuning to the telecom equipment vendor and/or operator. Several works by us and others have been carried out to study the effect of different SCTP parameters on the failover performance. The main contribution of this paper is that it gives a coherent treatment of how to configure the SCTP failover mechanism for carrier-grade telephony signaling, and provides practically usable configuration recommendations. The paper also discusses an alternate or complementary way of optimizing the SCTP failover mechanism by relaxing the exponential backoff that foregoes a retransmission timeout in SCTP. Some results showing significantly reduced failover times by use of this mechanism, with only marginal deteriorating effects on a signaling network, are discussed and analyzed in the paper.

SCTP congestion control includes the slow-start mechanism to probe the network for available bandwidth. In case of a path switch in a multihomed association, this mechanism may cause a sudden drop in throughput and increased message delays. By estimating the available bandwidth on the alternate path it is possible to utilize a more efficient startup scheme. In this paper, we analytically compare and quantify the degrading impact of slow start in relation to an ideal startup scheme. We consider three different scenarios where a path switch could occur. Further, we identify relevant traffic for these scenarios. Our results point out that the most prominent performance gain is seen for applications generating high traffic loads, like video conferencing. For this traffic, we have seen reductions in transfer time of more than 75% by an ideal startup scheme. Moreover, the results show an increasing impact of an improved startup mechanism with increasing RTTs