Company A has a hub-and spoke topology over an SP-managed infrastructure. To measure traffic performance metrics. IP SLA senders on all spoke CE routers and an IP SLA responder on the hub CE router. W

Memory usage on the hub is relatively stable because IP SLA responder session state is lightweight and does not grow significantly as more spokes are added.

Interface buffer utilization is driven by traffic volume and burst characteristics rather than directly by the count of IP SLA sessions, so it does not specifically reflect the scalability impact of adding spoke sites.

Each spoke independently sends probes to the hub and is not responsible for aggregating responses, so spoke CPU and memory usage does not scale with the total number of spoke sites in the topology.

The hub CE router runs the single IP SLA responder that handles probe responses for all spoke senders; as the number of spokes increases linearly, the per-packet CPU processing load on the hub scales proportionally, making hub CPU usage the primary scalability bottleneck to monitor.