One of the most critical challenges of the Internet of Things (IoT) is to provide real-time services. Therefore, to provide a secure, efficient, and stable communication platform in the Internet of Things, emerging architectures such as software-defined networks (SDN) being significant. This paper proposes a comprehensive SDN based fault-tolerant architecture in IoT environments. In the proposed scheme, a mathematical model called Shared Risk Link Group (SRLG) calculates redundant paths as the main and backup non-overlapping paths between network equipment. In addition to the fault tolerance (FT) discussion in the proposed scheme, service quality is considered in the proposed scheme. Putting the percentage of link usage and the rate of link delay in calculating link costs makes it possible to calculate two completely non-overlapping paths. The end-to-end delay and the degree of link congestion are minimal. We compare our proposed scheme with two policies for building routes from source to destination. The simulation results indicate that, while reducing the error recovery time, the proposed method leads to improved services quality parameters such as packet loss, delay, and packet jitter. The results show that in case of a link failure in the network, the recovery time in large scenarios is a maximum of 16 ms, which improved by 20%, compared to the disjoint paths (DP) method. Our Approach can decrease packet loss by approximately 30% compared to the Dijkstra’s algorithm and roughly 24% compared to the DP method. Also, SLRG reduces latency by approximately 36% compared to the Dijkstra’s algorithm and roughly 19% compared to the DP method. Last, the proposed scheme reduces Jitter by approximately 49% compared to the Dijkstra’s algorithm, and roughly 26% compared to the DP method.

物联网（IoT）的最关键挑战之一就是提供实时服务。因此，为了在物联网中提供安全，高效和稳定的通信平台，诸如软件定义网络（SDN）之类的新兴体系结构至关重要。本文提出了一种在物联网环境中基于SDN的全面容错架构。在提出的方案中，称为共享风险链接组（SRLG）的数学模型计算出冗余路径，作为网络设备之间的主要和备用非重叠路径。除了在提议的方案中讨论容错（FT）之外，在提议的方案中还考虑了服务质量。在计算链路成本时，将链路使用率百分比和链路延迟率放在一起，就可以计算两条完全不重叠的路径。端到端延迟和链路拥塞程度最小。我们将提议的方案与构建从源到目的地的路线的两种策略进行了比较。仿真结果表明，该方法在减少错误恢复时间的同时，还改善了服务质量参数，如丢包，时延和抖动等。结果表明，在网络中发生链路故障的情况下，与不相交路径（DP）方法相比，在大型方案中的恢复时间最多为16 ms，与之前相比，缩短了20％。与Dijkstra算法相比，我们的方法可以将数据包丢失减少大约30％，与DP方法相比，可以减少大约24％。而且，与Dijkstra的算法相比，SLRG的延迟减少了大约36％，与DP方法相比，减少了大约19％。持续，