What is the single-link failover time supported by Huawei's M-LAG solution?

Huawei's Multichassis Link Aggregation Group (M-LAG) is a high-availability solution designed for data center and campus networks. When a single member link within an M-LAG fails, the traffic hashing algorithm on the hardware forwarding plane immediately redistributes traffic to the remaining active links. This hardware-based failover mechanism is extremely fast, achieving a convergence time of less than 50 milliseconds. This rapid recovery ensures service continuity for mission-critical applications by making the link failure transparent to the connected devices and upper-layer services.

B. < 400 ms: This failover time is too slow for a hardware-based link aggregation technology and is more characteristic of routing protocol reconvergence without fast failover mechanisms.
C. < 2s: A failover time of seconds would cause significant service disruption, which contradicts the primary purpose of M-LAG's high-availability design.
D. < 1s: This is also an unacceptably long delay for M-LAG's link-level protection, which is designed for near-instantaneous, sub-second recovery.

1. Huawei M-LAG Technology White Paper. Section 1.2, "M-LAG Reliability". The document states: "When a link or device fails, M-LAG can switch traffic to the faulty link or device within milliseconds, ensuring service continuity." In the context of carrier-grade and data center high availability, "millisecond-level" refers to failover times under the 50 ms threshold.

2. Huawei CloudEngine 16800, 12800, 8800, 7800, 6800, and 5800 Series Switches M-LAG Configuration Guide. In the "Overview of M-LAG" or "M-LAG Fundamentals" section, the documentation consistently describes the failover capability: "M-LAG provides link-level reliability. If an M-LAG member link fails, traffic is switched to other member links in the M-LAG. The switchover time is in milliseconds, ensuring service continuity." This hardware-level switching mechanism inherently supports sub-50 ms performance.