A company deployed an Isilon cluster that consists of eight nodes connected to a main production subnet by all available 10GB ports. Later, the company decided to implement a disaster recovery solution with a separate replication subnet. During the reconfiguration, one of the 10Gb ports from each of the nodes 1-4 was moved into the new pool and SBR was activated.

After activating the replication policies, the administrator noticed that the main production ports on nodes 1-4 reached the bandwidth limit. What is a possible reason for this behavior?